ahong286286
ahong286286

ICML 2016 papers

Detailed Schedule

Monday – Neural Networks and Deep Learning

Session chair: Hugo Larochelle

Location:  Ballroom 1+2+Juliard
  • 10:20 – One-Shot Generalization in Deep Generative Models Danilo Rezende Google DeepMind, Shakir , Ivo Danihelka Google DeepMind, Karol Gregor DeepMind, Daan Wierstra Google DeepMindPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:37 – Learning to Generate with Memory Chongxuan Li Tsinghua University, Jun Zhu Tsinghua, Bo Zhang Tsinghua UniversityPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:54 – A Theory of Generative ConvNet Jianwen Xie UCLA, Yang Lu UCLA, Song-Chun Zhu UCLA, Yingnian Wu UCLAPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:30 – Deconstructing the Ladder Network Architecture Mohammad Pezeshki Universite de Montreal, Linxi Fan , Philemon Brakel , Aaron Courville , Yoshua Bengio U. of MontrealPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:47 – Normalization Propagation: A Parametric Technique for Removing Internal Covariate Shift in Deep Networks Devansh ArpitSUNY Buffalo, Yingbo Zhou SUNY Buffalo, Bhargava Kota SUNY Buffalo, Venu Govindaraju SUNY BuffaloPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 12:04 – Unitary Evolution Recurrent Neural Networks Martin Arjovsky University of Buenos Aires, Amar Shah University of Cambridge, Yoshua BengioPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract

    Session chair: Honglak Lee

    Location:  Ballroom 1+2+Juliard
  • 02:00 – Deep Speech 2 : End-to-End Speech Recognition in English and Mandarin Dario Amodei , Rishita Anubhai , Eric Battenberg , Carl Case , Jared Casper , Bryan Catanzaro , JingDong Chen , Mike Chrzanowski Baidu USA, Inc., Adam Coates , Greg Diamos Baidu USA, Inc., Erich Elsen Baidu USA, Inc., Jesse Engel , Linxi Fan , Christopher Fougner , Awni Hannun Baidu USA, Inc., Billy Jun , Tony Han , Patrick LeGresley , Xiangang Li Baidu, Libby Lin , Sharan Narang , Andrew Ng , Sherjil Ozair , Ryan Prenger , Sheng Qian Baidu, Jonathan Raiman , Sanjeev SatheeshBaidu SVAIL, David Seetapun , Shubho Sengupta , Chong Wang , Yi Wang , Zhiqian Wang , Bo Xiao , Yan Xie Baidu, Dani Yogatama , Jun Zhan , zhenyao ZhuPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 02:17 – Persistent RNNs: Stashing Recurrent Weights On-Chip Greg Diamos Baidu USA, Inc., Shubho Sengupta Baidu USA, Inc., Bryan Catanzaro Baidu USA, Inc., Mike Chrzanowski Baidu USA, Inc., Adam Coates , Erich Elsen Baidu USA, Inc., Jesse Engel Baidu USA, Inc., Awni Hannun Baidu USA, Inc., Sanjeev Satheesh Baidu USA, Inc.Paper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 02:34 – Sequence to Sequence Training of CTC-RNNs with Partial Windowing Kyuyeon Hwang Seoul National University, Wonyong SungSeoul National UniversityPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:10 – Analysis of Deep Neural Networks with Extended Data Jacobian Matrix Shengjie Wang University of Washington, Abdel-rahman Mohamed , Rich Caruana Microsoft, Jeff Bilmes U. of Washington, Matthai Plilipose , Matthew Richardson , Krzysztof Geras , Gregor Urban UC Irvine, Ozlem AslanPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:27 – Understanding and Improving Convolutional Neural Networks via Concatenated Rectified Linear Units Wenling Shang , Kihyuk Sohn NEC Laboratories America, Diogo Almeida Enlitic, Honglak Lee University of MichiganPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • Best paper award – 03:44 – Pixel Recurrent Neural Networks Aaron Van den Oord Google Deepmind, Nal Kalchbrenner Google Deepmind, Koray Kavukcuoglu Google DeepmindPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract

    ⤴

    Monday – Reinforcement Learning

    Session chair: Joelle Pineau

    Location:  Ballroom 3+4
  • 10:20 – Why Most Decisions Are Easy in Tetris—And Perhaps in Other Sequential Decision Problems, As Well Ozgur Simsek , Simon Algorta , Amit KothiyalPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:37 – Opponent Modeling in Deep Reinforcement Learning He He , Jordan , Kevin Kwok Massachusetts Institute of Technology, Hal Daume MarylandPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:54 – Control of Memory, Active Perception, and Action in Minecraft Junhyuk Oh University of Michigan, Valliappa ChockalingamUniversity of Michigan, Satinder , Honglak Lee University of MichiganPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:30 – Graying the black box: Understanding DQNs Tom Zahavy Technion, Nir Ben-Zrihem , Shie Mannor TechnionPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:47 – Benchmarking Deep Reinforcement Learning for Continuous Control Yan Duan University of California, Berk, Xi Chen University of California, Berkeley, Rein Houthooft Ghent University, John Schulman University of California, Berkeley, Pieter Abbeel BerkeleyPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • Best paper award – 12:04 – Dueling Network Architectures for Deep Reinforcement Learning Ziyu Wang Google Inc., Tom Schaul Google Inc., Matteo Hessel Google Deepmind, Hado van Hasselt Google DeepMind, Marc Lanctot Google Deepmind, Nando de Freitas University of OxfordPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract

    ⤴

    Monday – Optimization (Continuous)

    Session chair: Sebastien Bubeck

    Location:  Marquis
  • 10:20 – SDCA without Duality, Regularization, and Individual Convexity Shai Shalev-Shwartz Hebrew University of JerusalemPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:37 – Stochastic Variance Reduction for Nonconvex Optimization Sashank J. Reddi Carnegie Mellon University, Ahmed Hefny , Suvrit Sra, Barnabas Poczos , AlexPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:54 – Fast Rate Analysis of Some Stochastic Optimization Algorithms Chao Qu Nus, Huan Xu National University of Singapore, Chong jin Ong NusPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:30 – Black-box Optimization with a Politician Sebastien Bubeck Microsoft, Yin Tat Lee MITPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:47 – Starting Small – Learning with Adaptive Sample Sizes Hadi Daneshmand ETH Zurich, Aurelien Lucchi ETHZ, Thomas HofmannPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 12:04 – Primal-Dual Rates and Certificates Celestine Dünner IBM Research, Simone Forte Google, Martin Takac , Martin JaggiPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract

    ⤴

    Monday – Online Learning

    Session chair: Alina Beygelzimer

    Location:  Lyceum
  • 10:20 – Online Learning with Feedback Graphs Without the Graphs Alon Cohen Technion, Tamir Hazan , Tomer KorenPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:37 – Efficient Algorithms for Adversarial Contextual Learning Vasilis Syrgkanis Microsoft Research, Akshay Krishnamurthy Microsoft Research, Robert Schapire Microsoft ResearchPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:54 – BISTRO: An Efficient Relaxation-Based Method for Contextual Bandits Alexander Rakhlin , Karthik Sridharan CornellPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:30 – Online Stochastic Linear Optimization under One-bit Feedback Lijun Zhang Nanjing University, Tianbao Yang University of Iowa, Rong Jin Alibaba Group, Yichi Xiao Nanjing University, Zhi-hua ZhouPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:47 – Tracking Slowly Moving Clairvoyant: Optimal Dynamic Regret of Online Learning with True and Noisy Gradient Tianbao YangUniversity of Iowa, Lijun Zhang Nanjing University, Rong Jin , Jinfeng Yi IBM ResearchPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 12:04 – Adaptive Algorithms for Online Convex Optimization with Long-term Constraints Rodolphe Jenatton , Jim Huang Amazon, Cedric ArchambeauPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract

    Session chair: Csaba Szepesvari

    Location:  Liberty
  • 04:15 – Pricing a Low-regret Seller Hoda Heidari , Mohammad Mahdian Google, Umar Syed , Sergei Vassilvitskii , Sadra Yazdanbod Georgia Institute of TechnologyPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 04:32 – Multi-Player Bandits — a Musical Chairs Approach Jonathan Rosenski Weizmann Institute of Science, Ohad Shamir Weizmann Institute of Science, Liran Szlak Weizmann Institute of SciencePaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 04:49 – Contextual Combinatorial Cascading Bandits Shuai Li CUHK, Baoxiang Wang , Shengyu Zhang , Wei ChenPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:25 – Copeland Dueling Bandit Problem: Regret Lower Bound, Optimal Algorithm, and Computationally Efficient Algorithm Junpei Komiyama The University of Tokyo, Junya Honda The University of Tokyo, Hiroshi Nakagawa The University of TokyoPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:42 – DCM Bandits: Learning to Rank with Multiple Clicks Sumeet Katariya University of Wisconsin Madiso, Branislav Kveton Adobe Research, Csaba Szepesvari Alberta, Zheng WenPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:59 – Distributed Clustering of Linear Bandits in Peer to Peer Networks Nathan Korda University of Oxford, Balazs Szorenyi , Shuai LiUniversity of InsubriaPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract

    ⤴

    Monday – Clustering

    Session chair: Alexandru Niculescu-Mizil

    Location:  Empire
  • 10:20 – Correlation Clustering and Biclustering with Locally Bounded Errors Gregory Puleo UIUC, Olgica Milenkovic UIUCPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:37 –  K K-Means Clustering with Distributed Dimensions Hu Ding State University of New York at Buffalo, Yu Liu Tsinghua University, IIIS, Lingxiao Huang , Jian LiPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:54 – Speeding up k-means by approximating Euclidean distances via block vectors Thomas Bottesch Ulm University, Thomas BühlerAvira, Markus Kächele Ulm UniversityPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:30 – Fast k-means with accurate bounds James Newling Idiap Research Institute, Francois Fleuret Idiap research institutePaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:47 – k-variates++: more pluses in the k-means++ Richard Nock Nicta & ANU, Raphael Canyasse Ecole Polytechnique and The Technion, Roksana Boreli Data61, Frank Nielsen Ecole Polytechnique and Sony CS Labs Inc.Paper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 12:04 – Compressive Spectral Clustering Nicolas TREMBLAY INRIA Rennes, Gilles Puy Technicolor, Remi Gribonval INRIA, Pierre Vandergheynst EPFLPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract

    ⤴

    Monday – Bayesian Nonparametric Methods

    Session chair: Hal Daume III

    Location:  Soho
  • 10:20 – On collapsed representation of hierarchical Completely Random Measures Gaurav Pandey Indian Institute of Science, Ambedkar Dukkipati Indian Institute of SciencePaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:37 – Hawkes Processes with Stochastic Excitations Young Lee NICTA, Kar Wai Lim ANU, Cheng Soon Ong NICTAPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:54 – The Segmented iHMM: A Simple, Efficient Hierarchical Infinite HMM Ardavan Saeedi MIT, Matthew Hoffman Adobe Research, Matthew Johnson , Ryan Adams HarvardPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:30 – Markov Latent Feature Models Aonan Zhang Columbia University, John PaisleyPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:47 – Diversity-Promoting Bayesian Learning of Latent Variable Models Pengtao Xie Carnegie Mellon University, Jun Zhu Tsinghua, Eric Xing CMUPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 12:04 – Bayesian Poisson Tucker Decomposition for Learning the Structure of International Relations Aaron Schein , Mingyuan Zhou , Blei David Columbia, Hanna Wallach MicrosoftPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract

    ⤴

    Monday – Matrix Factorization / Neuroscience Applications

    Session chair: Jennifer Dy

    Location:  Liberty
  • 10:20 – The knockoff filter for FDR control in group-sparse and multitask regression Ran Dai The University of Chicago, Rina Barber The University of ChicagoPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:37 – A Simple and Provable Algorithm for Sparse Diagonal CCA Megasthenis Asteris University of Texas at Austin, Anastasios Kyrillidis , Oluwasanmi Koyejo Stanford University & University of Illinois at Urbana Champaign, Russell Poldrack Stanford UniversityPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:54 – Experimental Design on a Budget for Sparse Linear Models and Applications Sathya Narayanan Ravi UW Madison, Vamsi IthapuUW Madison, Sterling Johnson UW Madison, Vikas SinghPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:30 – Representational Similarity Learning with Application to Brain Networks Urvashi Oswal University of Wisconsin, Christopher CoxUniversity of Wisconsin, Matthew Lambon-Ralph , Timothy Rogers University of Wisconsin, Robert NowakPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:47 – Dictionary Learning for Massive Matrix Factorization Arthur Mensch Inria, Julien Mairal , Bertrand Thirion Inria, Gael VaroquauxPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 12:04 – A Random Matrix Approach to Echo-State Neural Networks Romain Couillet CentraleSupelec, Gilles Wainrib ENS Ulm, Paris, France, Hafiz Tiomoko Ali CentraleSupelec, Gif-sur-Yvette, France, Harry Sevi ENS Lyon, Lyon, ParisPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract

    ⤴

    Monday – Optimization / Online Learning

    Session chair: Satyen Kale

    Location:  Ballroom 3+4
  • 02:00 – Shifting Regret, Mirror Descent, and Matrices Andras Gyorgy , Csaba Szepesvari AlbertaPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 02:17 – Heteroscedastic Sequences: Beyond Gaussianity Oren , Shie Mannor TechnionPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 02:34 – Convergence of Stochastic Gradient Descent for PCA Ohad Shamir Weizmann Institute of SciencePaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:10 – Fast Stochastic Algorithms for SVD and PCA: Convergence Properties and Convexity Ohad Shamir Weizmann Institute of SciencePaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:27 – Faster Eigenvector Computation via Shift-and-Invert Preconditioning Dan Garber TTI Chicago, Elad Hazan Princeton University, Chi Jin UC Berkeley, Sham , Cameron Musco Massachusetts Institute of Technology, Praneeth Netrapalli Microsoft Research, Aaron SidfordMicrosoft Research, New EnglandPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:44 – Solving Ridge Regression using Sketched Preconditioned SVRG Alon Gonen Hebrew University of Jerusalem, Francesco OrabonaYahoo, Shai Shalev-Shwartz Hebrew University of JerusalemPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract

    ⤴

    Monday – Machine Learning Applications

    Session chair: Balazs Kegl

    Location:  Marquis
  • 02:00 – Bounded Off-Policy Evaluation with Missing Data for Course Recommendation and Curriculum Design William Hoiles UCLA, Mihaela van der SchaarPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 02:17 – Dealbreaker: A Nonlinear Latent Variable Model for Educational Data Andrew Lan Rice University, Tom Goldstein University of Maryland, Richard Baraniuk Rice University, Christoph Studer Cornell UniversityPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 02:34 – Estimating Cosmological Parameters from the Dark Matter Distribution Siamak Ravanbakhsh CMU, Junier Oliva CMU, Sebastian Fromenteau Carnegie Mellon Unitversity, Layne Price Carnegie Mellon Unitversity, Shirley Ho Carnegie Mellon Unitversity, Jeff Schneider Carnegie Mellon Unitversity, Barnabas PoczosPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:10 – BASC: Applying Bayesian Optimization to the Search for Global Minima on Potential Energy Surfaces Shane Carr Washington University in St. L, Roman Garnett Wustl, Cynthia Lo Washington University in St. LouisPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:27 – Predictive Entropy Search for Multi-objective Bayesian Optimization Daniel Hernandez-Lobato , Jose miguel Hernandez-Lobato , Amar Shah University of Cambridge, Ryan Adams HarvardPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:44 – Pareto Frontier Learning with Expensive Correlated Objectives Amar Shah University of Cambridge, ZoubinPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract

    ⤴

    Monday – Matrix Factorization and Related Topics

    Session chair: Alex Kulesza

    Location:  Lyceum
  • 02:00 – Complex Embeddings for Simple Link Prediction Théo Trouillon Xerox, Johannes Welbl , Sebastian Riedel , Eric Gaussier , Guillaume BouchardPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 02:17 – PAC learning of Probabilistic Automaton based on the Method of Moments Hadrien Glaude University of Lille, Olivier Pietquin Univ. Lille, CRIStAL, UMR 9189, SequeL Team, Villeneuve d’Ascq, 59650, FRANCEPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 02:34 – Rich Component Analysis Rong Ge , James ZouPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:10 – Beyond CCA: Moment Matching for Multi-View Models Anastasia Podosinnikova INRIA – ENS, Francis Bach Inria, Simon Lacoste-Julien INRIAPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:27 – Isotonic Hawkes Processes Yichen Wang Georgia Tech, Bo Xie , Nan Du , Le Song GatechPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:44 – Non-negative Matrix Factorization under Heavy Noise Chiranjib Bhattacharya , Navin Goyal Microsoft Research India, Ravindran Kannan Microsoft Reseach India, Jagdeep Pani Indian Institute of SciencePaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract

    ⤴

    Monday – Bandit Problems

    Session chair: Shie Mannor

    Location:  Empire
  • 02:00 – An optimal algorithm for the Thresholding Bandit Problem Andrea LOCATELLI University of Potsdam, Maurilio Gutzeit Universität Potsdam, Alexandra CarpentierPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 02:17 – Anytime Exploration for Multi-armed Bandits using Confidence Information Kwang-Sung Jun UW-Madison, Robert NowakPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 02:34 – Anytime optimal algorithms in stochastic multi-armed bandits Rémy Degenne Université Paris Diderot, Vianney PerchetPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:10 – PAC Lower Bounds and Efficient Algorithms for The Max  K K-Armed Bandit Problem Yahel David Technion, Nahum ShimkinTechnionPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:27 – Conservative Bandits Yifan Wu , Roshan Shariff University of Alberta, Tor , Csaba Szepesvari AlbertaPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:44 – No-Regret Algorithms for Heavy-Tailed Linear Bandits Andres Munoz Medina , Scott YangPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract

    ⤴

    Monday – Graphical Models

    Session chair: Stefano Ermon

    Location:  Soho
  • 02:00 – Hierarchical Span-Based Conditional Random Fields for Labeling and Segmenting Events in Wearable Sensor Data Streams Roy Adams Univ. of Massachusetts-Amherst, Nazir Saleheen , Edison Thomaz , Abhinav Parate , Santosh Kumar , Benjamin MarlinPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 02:17 – Efficient Multi-Instance Learning for Activity Recognition from Time Series Data Using an Auto-Regressive Hidden Markov Model Xinze Guan Oregon State University, Raviv Raich Oregon State University, Weng-keenPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 02:34 – Clustering High Dimensional Categorical Data via Topographical Features Chao Chen CUNY, Novi Quadrianto University of SussexPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:10 – Nonlinear Statistical Learning with Truncated Gaussian Graphical Models Qinliang Su Duke University, xuejun Liao , changyou Chen , Lawrence CarinPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:27 – Collapsed Variational Inference for Sum-Product Networks Han Zhao Carnegie Mellon University, Tameem Adel University of Amsterdam, Geoff Gordon CMU, Brandon Amos Carnegie Mellon UniversityPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:44 – Square Root Graphical Models: Multivariate Generalizations of Univariate Exponential Families that Permit Positive Dependencies David Inouye University of Texas at Austin, Pradeep Ravikumar UT Austin, InderjitPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract

    ⤴

    Monday – Transfer Learning / Learning Theory

    Session chair: Alina Beygelzimer

    Location:  Liberty
  • 02:00 – A New PAC-Bayesian Perspective on Domain Adaptation Pascal Germain INRIA, Amaury Habrard , François Laviolette GRAAL, Université Laval, Emilie MorvantPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 02:17 – Domain Adaptation with Conditional Transferable Components Mingming Gong University of Technology Sydne, Kun ZhangCarnegie Mellon University, Tongliang Liu MPI Tuebingen, Dacheng Tao , Clark Glymour , Bernhard SchölkopfPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 02:34 – Minimizing the Maximal Loss: How and Why Shai Shalev-Shwartz Hebrew University of Jerusalem, Yonatan Wexler OrcamPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 03:10 – Accurate Robust and Efficient Error Estimation for Decision Trees Lixin Fan Nokia TechnologiesPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:27 – The Teaching Dimension of Linear Learners Ji Liu University of Rochester, Xiaojin Zhu University of Wisconsin, Hrag OhannessianUniversity of Wisconsin-MadisonPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 03:44 – Loss factorization, weakly supervised learning and label noise robustness Giorgio Patrini ANU / Data61, Frank Nielsen Ecole Polytechnique and Sony CS Labs Inc., Richard Nock Nicta & ANU, Marcello Carioni Max Planck InstitutePaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract

    ⤴

    Monday – Neural Networks and Deep Learning I

    Session chair: Kyunghyun Cho

    Location:  Ballroom 1+2+Juliard
  • 04:15 – Texture Networks: Feed-forward Synthesis of Textures and Stylized Images Dmitry Ulyanov Skolkovo institute of science , Vadim Lebedev , Andrea , Victor Lempitsky Skolkovo Institute of Sci&TechPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 04:32 – Discrete Deep Feature Extraction: A Theory and New Architectures Thomas Wiatowski ETH Zurich, Michael Tschannen ETH Zurich, Aleksandar Stanic ETH Zurich, Philipp Grohs University of Vienna, Helmut Boelcskei ETH ZurichPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 04:49 – Deep Structured Energy Based Models for Anomaly Detection Shuangfei Zhai Binghamton University, Yu Cheng IBM Research, Weining Lu Tsinghua Univerisity, Zhongfei Zhang Binghamton UniversityPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:25 – Noisy Activation Functions Caglar Gülçehre , Marcin Moczulski , Misha Denil , Yoshua Bengio U. of MontrealPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:42 – A Kronecker-factored approximate Fisher matrix for convolution layers Roger Grosse , James Martens University of TorontoPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:59 – Recurrent Orthogonal Networks and Long-Memory Tasks Mikael Henaff NYU, Facebook, Arthur , YannPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract

    ⤴

    Monday – Neural Networks and Deep Learning II (Computer Vision)

    Session chair: Laurens van der Maaten

    Location:  Ballroom 3+4
  • 04:15 – Group Equivariant Convolutional Networks Taco Cohen University of Amsterdam, Max Welling University of Amsterdam / CIFARPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 04:32 – Learning End-to-end Video Classification with Rank-Pooling Basura Fernando ANU Canberra Australia, Stephen Gould Australian National UniversityPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 04:49 – Learning Physical Intuition of Block Towers by Example Adam Lerer Facebook AI Research, Sam Gross Facebook AI Research, Rob Fergus Facebook AI ResearchPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:25 – Large-Margin Softmax Loss for Convolutional Neural Networks Weiyang Liu Peking University, Yandong Wen South China University of Technology, Zhiding Yu Carnegie Mellon University, Meng Yang Shenzhen UniversityPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:42 – Network Morphism Tao Wei University at Buffalo, Changhu Wang Microsoft Research, Yong Rui Microsoft Research, Chang Wen ChenPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:59 – Multi-Bias Non-linear Activation in Deep Neural Networks Hongyang Li The Chinese Univ. of HK, Wanli Ouyang , Xiaogang WangPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract

    ⤴

    Monday – Approximate Inference

    Session chair: Jun Zhu

    Location:  Marquis
  • 04:15 – Boolean Matrix Factorization and Noisy Completion via Message Passing Siamak Ravanbakhsh CMU, Barnabas Poczos , Russell Greiner University of AlbertaPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 04:32 – Stochastic Discrete Clenshaw-Curtis Quadrature Nico Piatkowski TU Dortmund, Katharina Morik TU DortmundPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 04:49 – Beyond Parity Constraints: Fourier Analysis of Hash Functions for Inference Tudor Achim Stanford University, Ashish SabharwalAllen Institute for AI, Stefano ErmonPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:25 – Variable Elimination in the Fourier Domain Yexiang Xue Cornell University, Stefano Ermon , Ronan Le Bras Cornell University, Carla , BartPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:42 – Learning and Inference via Maximum Inner Product Search Stephen Mussmann Stanford University, Stefano ErmonPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:59 – Analysis of Variational Bayesian Factorizations for Sparse and Low-Rank Estimation David Wipf Microsoft Research, BeijingPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract

    ⤴

    Monday – Metric and Manifold Learning / Kernel Methods

    Session chair: Le Song

    Location:  Lyceum
  • 04:15 – Fast k-Nearest Neighbour Search via Dynamic Continuous Indexing Ke Li UC Berkeley, Jitendra MalikPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 04:32 – Geometric Mean Metric Learning Pourya Zadeh Tehran university, Reshad Hosseini , Suvrit SraPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 04:49 – Low-rank tensor completion: a Riemannian manifold preconditioning approach Hiroyuki Kasai The University of Electro-Comm, Bamdev Mishra Amazon Development Centre IndiaPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:25 – The Variational Nystrom method for large-scale spectral problems Max Vladymyrov Yahoo Labs, Miguel Carreira-Perpinan UC MercedPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:42 – Fast DPP Sampling for Nystro ̈m with Application to Kernel Methods Chengtao Li MIT, Stefanie Jegelka MIT, Suvrit SraPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:59 – Computationally Efficient Nystr\”{o}m Approximation using Fast Transforms Si Si , Cho-Jui Hsieh UC Davis, InderjitPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract

    ⤴

    Monday – Statistical Learning Theory

    Session chair: Ohad Shamir

    Location:  Empire
  • 04:15 – Barron and Cover’s Theory in Supervised Learning and its Application to Lasso Masanori Kawakita Kyushu University, Jun’ichi Takeuchi Kyushu UniversityPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 04:32 – Exact Exponent in Optimal Rates for Crowdsourcing Chao Gao Yale University, Yu Lu Yale University, Dengyong Zhou Microsoft ResearchPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 04:49 – Generalization Properties and Implicit Regularization for Multiple Passes SGM Junhong Lin Istituto Italiano di Tecnologi, Raffaello Camoriano IIT Italy and UNIGE Italy, Lorenzo Rosasco Istituto Italiano di Tecnologia, Università degli Studi di Genova and MITPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:25 – Generalized Direct Change Estimation in Ising Model Structure Farideh Fazayeli University of Minnesota, Arindam BanerjeePaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:42 – Gaussian process nonparametric tensor estimator and its minimax optimality Heishiro Kanagawa Tokyo Institute of Technology, Taiji Suzuki , Hayato Kobayashi Yahoo Japan Corporation, Nobuyuki Shimizu , Yukihiro Tagami Yahoo Japan CorporationPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:59 – Minimum Regret Search for Single- and Multi-Task Optimization Jan Hendrik Metzen University BremenPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract

    ⤴

    Monday – Structured Prediction / Monte Carlo Methods

    Session chair: Martin Jaggi

    Location:  Soho
  • 04:15 – The Sum-Product Theorem: A Foundation for Learning Tractable Models Abram Friesen University of Washington, PedroPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 04:32 – Train and Test Tightness of LP Relaxations in Structured Prediction Ofer Meshi , Mehrdad Mahdavi , Adrian Weller University of Cambridge, David Sontag NYUPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 04:49 – Evasion and Hardening of Tree Ensemble Classifiers Alex Kantchelian University of California, Berk, J. D. Tygar UC Berkeley, Anthony Joseph UC BerkeleyPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:25 – Importance Sampling Tree for Large-scale Empirical Expectation Olivier CANEVET Idiap Research Institut, Cijo Jose Idiap Research Institute/ Éc , Francois Fleuret Idiap research institutePaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:42 – Stratified Sampling Meets Machine Learning Edo Liberty , Kevin Lang Yahoo Labs, Konstantin Shmakov Yahoo LabsPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract
  • 05:59 – Scalable Discrete Sampling as a Multi-Armed Bandit Problem Yutian Chen University of Cambridge, ZoubinPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract

    ⤴

    Tuesday – Neural Networks and Deep Learning

    Session chair: Hal Daume III

    Location:  Ballroom 1+2+Juliard
  • 10:30 – Strongly-Typed Recurrent Neural Networks David Balduzzi , Muhammad Ghifary Victoria University Wellington, Weta DigitalPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:47 – A Convolutional Attention Network for Extreme Summarization of Source Code Miltiadis Allamanis University of Edinburgh, UK, Hao Peng Peking University, China, Charles SuttonPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:04 – Ask Me Anything: Dynamic Memory Networks for Natural Language Processing Ankit Kumar MetaMind, Ozan Irsoy MetaMind, Peter Ondruska , Mohit Iyyer MetaMind, James Bradbury MetaMind, Ishaan Gulrajani MetaMind, Victor Zhong MetaMind, Romain Paulus MetaMind, Richard SocherPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:40 – Dynamic Memory Networks for Visual and Textual Question Answering Caiming Xiong MetaMind, Stephen Merity MetaMind, Richard SocherPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:57 – Supervised and Semi-Supervised Text Categorization using LSTM for Region Embeddings Rie Johnson RJ Research Consulting, Tong ZhangPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 12:14 – PHOG: Probabilistic Model for Code Pavol Bielik ETH Zurich, Veselin Raychev ETH Zurich, Martin Vechev ETH ZurichPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract

    ⤴

    Tuesday – Reinforcement Learning

    Session chair: Tom Erez

    Location:  Ballroom 3+4
  • 10:30 – On the Analysis of Complex Backup Strategies in Monte Carlo Tree Search Piyush Khandelwal University of Texas at Austin, Elad Liebman , Scott , Peter Stone U. of TexasPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:47 – Generalization and Exploration via Randomized Value Functions Ian Osband Stanford, Benjamin Van Roy Stanford University, Zheng Wen Adobe ResearchPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:04 – Doubly Robust Off-policy Value Evaluation for Reinforcement Learning Nan Jiang University of Michigan, Lihong Li MicrosoftPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:40 – Near Optimal Behavior via Approximate State Abstraction David Abel Brown University, David Hershkowitz Brown University, Michael LittmanPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:57 – Model-Free Trajectory Optimization\\ for Reinforcement Learning Riad Akrour TU Darmstadt, Gerhard Neumann , Hany Abdulsamad, Abbas AbdolmalekiPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 12:14 – Model-Free Imitation Learning with Policy Optimization Jonathan Ho Stanford, Jayesh Gupta Stanford University, Stefano ErmonPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract

    Session chair: Lihong Li

    Location:  Marquis
  • 03:40 – Guided Cost Learning: Deep Inverse Optimal Control via Policy Optimization Chelsea Finn UC Berkeley, Sergey Levine , Pieter Abbeel BerkeleyPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 03:57 – Smooth Imitation Learning for Online Sequence Prediction Hoang Le Caltech, Andrew Kang , Yisong Yue Caltech, Peter CarrPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:14 – Continuous Deep Q-Learning with Model-based Acceleration Shixiang Gu University of Cambridge, Timothy Lillicrap Google DeepMind, Ilya Sutskever OpenAI, Sergey Levine GooglePaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:31 – Asynchronous Methods for Deep Reinforcement Learning Volodymyr Mnih Google DeepMind, Adria Puigdomenech Badia Google DeepMind, Mehdi Mirza , Alex Graves Google DeepMind, Timothy Lillicrap Google DeepMind, Tim Harley Google DeepMind, David Silver Google Deepmind, Koray Kavukcuoglu Google DeepmindPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    Session chair: Tom Erez

    Location:  Marquis
  • 05:10 – Estimating Maximum Expected Value through Gaussian Approximation Carlo D’Eramo Politecnico di Milano, Marcello RestelliPolitecnico di Milano, Alessandro Nuara Politecnico di MilanoPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 05:27 – Data-Efficient Off-Policy Policy Evaluation for Reinforcement Learning Philip Thomas CMU, EmmaPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 05:44 – Cumulative Prospect Theory Meets Reinforcement Learning: Prediction and Control Prashanth L.A. University of Maryland , Cheng Jie University of Maryland – College Park, Michael Fu University of Maryland – College Park, Steve Marcus University of Maryland – College Park, Csaba Szepesvari AlbertaPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 06:01 – Softened Approximate Policy Iteration for Markov Games Julien Pérolat Univ. Lille, Bilal Piot Univ. Lille, Matthieu Geist , Bruno Scherrer , Olivier Pietquin Univ. Lille, CRIStAL, UMR 9189, SequeL Team, Villeneuve d’Ascq, 59650, FRANCEPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    ⤴

    Tuesday – Optimization (Combinatorial)

    Session chair: Andreas Krause

    Location:  Marquis
  • 10:30 – Algorithms for Optimizing the Ratio of Submodular Functions Wenruo Bai University of Washington, Rishabh Iyer , Kai Wei , Jeff Bilmes U. of WashingtonPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:47 – Horizontally Scalable Submodular Maximization Mario Lucic ETH Zurich, Olivier Bachem ETH Zurich, Morteza ZadimoghaddamGoogle Research, Andreas KrausePaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:04 – Learning Sparse Combinatorial Representations via Two-stage Submodular Maximization Eric Balkanski , Baharan MirzasoleimanETH Zurich, Andreas Krause , Yaron SingerPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:40 – Fast Constrained Submodular Maximization: Personalized Data Summarization Baharan Mirzasoleiman ETH Zurich, Ashwinkumar Badanidiyuru Google Research, Amin Karbasi Yale UniversityPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:57 – A Box-Constrained Approach for Hard Permutation Problems Cong Han Lim Univ of Wisconsin – Madison, Steve Wright University of Wisconsin-MadisonPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 12:14 – A Convex Atomic-Norm Approach to Multiple Sequence Alignment and Motif Discovery Ian En-Hsu Yen University of Texas at Austin, Xin Lin University of Texas at Austin, Jiong Zhang University of Texas at Austin, Pradeep Ravikumar UT Austin, InderjitPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract

    ⤴

    Tuesday – Unsupervised Learning / Representation Learning

    Session chair: Jennifer Dy

    Location:  Lyceum
  • 10:30 – Nonparametric Canonical Correlation Analysis Tomer Michaeli Technion, Weiran Wang , Karen Livescu TTI ChicagoPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:47 – The Information Sieve Greg Ver Steeg Information Sciences Institute, Aram Galstyan Information Sciences InstitutePaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:04 – Gromov-Wasserstein Averaging of Kernel and Distance Matrices Gabriel Peyré , Marco Cuturi Kyoto, Justin SolomonPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:40 – Learning Representations for Counterfactual Inference Fredrik Johansson Chalmers Technical University, Uri Shalit , David SontagNYUPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:57 – Why Regularized Auto-Encoders learn Sparse Representation? Devansh Arpit SUNY Buffalo, Yingbo Zhou SUNY Buffalo, Hung NgoSUNY Buffalo, Venu Govindaraju SUNY BuffaloPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 12:14 – Robust Random Cut Forest Based Anomaly Detection on Streams Sudipto Guha University of Pennsylvania, Nina Mishra , Gourav Roy Amazon, Okke Schrijvers Stanford UniversityPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract

    ⤴

    Tuesday – Sampling / Kernel Methods

    Session chair: Marius Kloft

    Location:  Empire
  • 10:30 – Mixing Rates for the Alternating Gibbs Sampler over Restricted Boltzmann Machines and Friends Christopher Tosh UC San DiegoPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:47 – Pliable Rejection Sampling Akram Erraqabi Inria Lille Nord Europe, Michal Valko Inria Lille – Nord Europe, Alexandra Carpentier , Odalric Maillard InriaPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:04 – A Kernel Test of Goodness of Fit Kacper.chwialkowski@ Chwialkowski Ucl, Heiko Strathmann University College London, Arthur Gretton UCL GatsbyPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:40 – A Kernelized Stein Discrepancy for Goodness-of-fit Tests qiang liu , Jason Lee UC Berkeley, MichaelPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:57 – Additive Approximations in High Dimensional Nonparametric Regression via the SALSA Kirthevasan Kandasamy Carnegie Mellon University, Yaoliang YuPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 12:14 – Doubly Decomposing Nonparametric Tensor Regression Masaaki Imaizumi University of Tokyo, Kohei Hayashi NIIPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract

    ⤴

    Tuesday – Sparsity and Compressed Sensing

    Session chair: Gal Chechik

    Location:  Soho
  • 10:30 – The Information-Theoretic Requirements of Subspace Clustering with Missing Data Daniel Pimentel-Alarcon UW-Madison, Robert Nowak UW MadisonPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:47 – Robust Principal Component Analysis with Side Information Kai-Yang Chiang UT Austin, Cho-Jui Hsieh UC Davis, InderjitPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:04 – Online Low-Rank Subspace Clustering by Basis Dictionary Pursuit Jie Shen Rutgers University, Ping Li Rutgers, Huan XuPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:40 – Provable Non-convex Phase Retrieval with Outliers: Median TruncatedWirtinger Flow Huishuai Zhang Syracuse University, Yuejie Chi Ohio State University, Yingbin Liang Syracuse UniversityPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:57 – Estimating Structured Vector Autoregressive Models Igor Melnyk University of Minnesota, Arindam BanerjeePaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 12:14 – Towards Faster Rates and Oracle Property for Low-Rank Matrix Estimation Huan Gui University of Illinois at Urba, Jiawei Hanuniversity of illinois at urbana-champaign, Quanquan GuPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract

    ⤴

    Tuesday – Approximate Inference

    Session chair: Jon Mcauliffe

    Location:  Liberty
  • 10:30 – Hierarchical Variational Models Rajesh Ranganath , Dustin Tran Columbia University, Blei David ColumbiaPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 10:47 – A Variational Analysis of Stochastic Gradient Algorithms Stephan Mandt Columbia University, Matthew Hoffman Adobe Research, Blei David ColumbiaPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:04 – Black-Box Alpha Divergence Minimization Jose miguel Hernandez-Lobato , Yingzhen Li University of Cambridge, Mark RowlandUniversity of Cambridge, Thang Bui University of Cambridge, Daniel Hernandez-Lobato , Richard Turner University of CambridgePaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:40 – Variational Inference for Monte Carlo Objectives Andriy Mnih , Danilo Rezende Google DeepMindPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 11:57 – Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning Yarin Gal University of Cambridge, ZoubinPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract
  • 12:14 – Auxiliary Deep Generative Models Lars Maaløe Technical University Denmark, Casper Kaae Sønderby University of Copenhagen, Søren Kaae Sønderby University of Copenagen, Ole Winther Technical University of DenmarkPaper | Reviews | Rebuttal | Poster session on monday afternoon 3:00pm-7:00pm | Abstract

    ⤴

    Tuesday – Neural Networks and Deep Learning I

    Session chair: Nicolas Le Roux

    Location:  Ballroom 1+2+Juliard
  • 03:40 – Factored Temporal Sigmoid Belief Networks for Sequence Learning Jiaming Song Tsinghua University, Zhe Gan Duke University, Lawrence CarinPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 03:57 – Bidirectional Helmholtz Machines Jorg Bornschein University of Montreal, Samira Shabanian University of Montreal, Asja Fischer , Yoshua Bengio U. of MontrealPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:14 – Structured and Efficient Variational Deep Learning with Matrix Gaussian Posteriors Christos Louizos University of Amsterdam, Max Welling University of Amsterdam / CIFARPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:31 – Dropout distillation Samuel Rota Bulò FBK, Lorenzo Porzi FBK, Peter Kontschieder Microsoft Research CambridgePaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    Session chair: Yoshua Bengio

    Location:  Ballroom 1+2+Juliard
  • 05:10 – Expressiveness of Rectifier Networks Xingyuan Pan University of Utah, Vivek Srikumar University of UtahPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 05:27 – Convolutional Rectifier Networks as Generalized Tensor Decompositions Nadav Cohen Hebrew University of Jerusalem, Amnon Shashua MobileyePaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 05:44 – Fixed Point Quantization of Deep Convolutional Networks Darryl Lin Qualcomm Research, Sachin Talathi Qualcomm Research, Sreekanth Annapureddy NetraDyne Inc.Paper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 06:01 – CryptoNets: Applying Neural Networks to Encrypted Data with High Throughput and Accuracy Ran Gilad-Bachrach Microsoft Research, Nathan Dowlin Princeton, Kim Laine Microsoft Research, Kristin Lauter Microsoft Research, Michael Naehrig Microsoft Research, John Wernsing Microsoft ResearchPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    ⤴

    Tuesday – Neural Networks and Deep Learning II

    Session chair: Alexandru Niculescu-Mizil

    Location:  Ballroom 3+4
  • 03:40 – Revisiting Semi-Supervised Learning with Graph Embeddings Zhilin Yang Carnegie Mellon University, William Cohen CMU, Ruslan Salakhudinov U. of TorontoPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 03:57 – ADIOS: Architectures Deep In Output Space Moustapha Cisse , Maruan Al-Shedivat CMU, Samy Bengio GooglePaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:14 – Unsupervised Deep Embedding for Clustering Analysis Junyuan Xie University of Washington, Ross Girshick Facebook, Ali FarhadiUniversity of WashingtonPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:31 – Learning Convolutional Neural Networks for Graphs Mathias Niepert NEC Laboratories Europe, Mohamed Ahmed NEC Laboratories Europe, Konstantin Kutzkov NEC Laboratories EuropePaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    Session chair: David Sontag

    Location:  Ballroom 3+4
  • 05:10 – Correcting Forecasts with Multifactor Neural Attention Matthew Riemer IBM, Aditya Vempaty IBM, Flavio Calmon IBM, Fenno HeathIBM, Richard Hull IBM, Elham Khabiri IBMPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 05:27 – Meta-Learning with Memory-Augmented Neural Networks Adam Santoro Google DeepMind, Sergey Bartunov Higher School of Economics, Matthew Botvinick , Daan Wierstra Google DeepMind, Timothy Lillicrap Google DeepMindPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 05:44 – Learning Simple Algorithms from Examples Wojciech Zaremba , Tomas , Armand Joulin , Rob Fergus Facebook AI ResearchPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 06:01 – Associative Long Short-Term Memory Ivo Danihelka Google DeepMind, Greg Wayne Google DeepMind, Benigno Uria Google DeepMind, Nal Kalchbrenner Google Deepmind, Alex Graves Google DeepMindPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    ⤴

    Tuesday – Optimization (Continuous)

    Session chair: Le Song

    Location:  Lyceum
  • 03:40 – On the Statistical Limits of Convex Relaxations Zhaoran Wang Princeton University, Quanquan Gu , HanPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 03:57 – Faster Convex Optimization: Simulated Annealing with an Efficient Universal Barrier Jacob Abernethy U. of Michigan, Elad HazanPrinceton UniversityPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:14 – A ranking approach to global optimization Cedric Malherbe ENS Cachan, Emile Contal ENS Cachan, Nicolas Vayatis ENS CachanPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:31 – Epigraph projections for fast general convex programming Po-Wei Wang Carnegie Mellon University, Matt Wytock , ZicoPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    Session chair: Tong Zhang

    Location:  Lyceum
  • 05:10 – Low-rank Solutions of Linear Matrix Equations via Procrustes Flow Stephen Tu UC Berkeley, Ross Boczar UC Berkeley, Max Simchowitz UC Berkeley, mahdi Soltanolkotabi , Ben Recht BerkeleyPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 05:27 – Quadratic Optimization with Orthogonality Constraints: Explicit Lojasiewicz Exponent and Linear Convergence of Line-Search Methods Huikang Liu CUHK, Weijie Wu CUHK, Anthony Man-Cho SoPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 05:44 – Efficient Algorithms for Large-scale Generalized Eigenvector Computation and Canonical Correlation Analysis Rong Ge , Chi JinUC Berkeley, Sham , Praneeth Netrapalli Microsoft Research, Aaron Sidford Microsoft Research, New EnglandPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 06:01 – Matrix Eigen-decomposition via Doubly Stochastic Riemannian Optimization Zhiqiang Xu Institute of Infocomm Research, Peilin Zhao I2R, ASTAR, Jianneng Cao , Xiaoli LiPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    ⤴

    Tuesday – Matrix Factorization and Related Topics

    Session chair: Roger Grosse

    Location:  Empire
  • 03:40 – Principal Component Projection Without Principal Component Analysis Roy Frostig Stanford University, Cameron MuscoMassachusetts Institute of Technology, Christopher Musco Mass. Institute of Technology, Aaron Sidford Microsoft Research, New EnglandPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 03:57 – Recovery guarantee of weighted low-rank approximation via alternating minimization Yuanzhi Li Princeton University, Yingyu LiangPrinceton, Andrej Risteski Princeton UniversityPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:14 – Tensor Decomposition via Joint Matrix Schur Decomposition Nicolo Colombo Univ of Luxembourg, Nikos Vlassis AdobePaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:31 – Fast methods for estimating the Numerical rank of large matrices Shashanka Ubaru University of Minnesota, Yousef Saad University of MinnesotaPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    ⤴

    Tuesday – Unsupervised Learning / Applications

    Session chair: Jeff Bilmes

    Location:  Soho
  • 03:40 – Markov-modulated Marked Poisson Processes for Check-in Data Jiangwei Pan Duke University, Vinayak Rao Purdue University, Pankaj Agarwal Duke Univeristy, Alan Gelfand Duke UniversityPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 03:57 – Hierarchical Compound Poisson Factorization Mehmet Basbug Princeton University, Barbara EngelhardtPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:14 – Dirichlet Process Mixture Model for Correcting Technical Variation in Single-Cell Gene Expression Data Sandhya PrabhakaranColumbia University, Elham Azizi Columbia University, Ambrose Carr Columbia University, Dana Pe’er Columbia UniversityPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:31 – Automatic Construction of Nonparametric Relational Regression Models for Multiple Time Series Yunseong Hwang UNIST, Anh Tong UNIST, Jaesik Choi UNISTPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    ⤴

    Tuesday – Learning Theory

    Session chair: Jon Mcauliffe

    Location:  Liberty
  • 03:40 – Truthful Univariate Estimators Ioannis Caragiannis University of Patras, Ariel Procaccia Carnegie Mellon University, Nisarg ShahCarnegie Mellon UniversityPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 03:57 – Fast Algorithms for Segmented Regression Jayadev Acharya MIT, Ilias Diakonikolas , Jerry Li MIT, Ludwig SchmidtPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:14 – Stochastically Transitive Models for Pairwise Comparisons: Statistical and Computational Issues Nihar Shah UC Berkeley, Sivaraman Balakrishnan CMU, Aditya Guntuboyina UC Berkeley, Martin Wainwright UC BerkeleyPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:31 – Provable Algorithms for Inference in Topic Models Sanjeev Arora Princeton University, Rong Ge , Frederic Koehler Princeton University, Tengyu Ma Princeton University, Ankur MoitraPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    ⤴

    Tuesday – Large Scale Learning and Big Data

    Session chair: Andreas Krause

    Location:  Empire
  • 05:10 – Extreme F-measure Maximization using Sparse Probability Estimates Kalina Jasinska , Krzysztof Dembczynski , Robert Busa-Fekete , Karlson Pfannschmidt , Timo Klerx , Eyke HullermeierPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 05:27 – Stochastic Optimization for Multiview Representation Learning using Partial Least Squares Raman Arora Johns Hopkins University, Poorya Mianjy Johns Hopkins University, Teodor MarinovPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 05:44 – Gaussian quadrature for matrix inverse forms with applications Chengtao Li MIT, Suvrit Sra , Stefanie Jegelka MITPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 06:01 – A Subspace Learning Approach for High Dimensional Matrix Decomposition with Efficient Column/Row Sampling Mostafa Rahmani University of Central Florida, Geroge Atia University of Central FloridaPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    ⤴

    Tuesday – Graphical Models

    Session chair: Jun Zhu

    Location:  Soho
  • 05:10 – Uprooting and Rerooting Graphical Models Adrian Weller University of CambridgePaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 05:27 – Structure Learning of Partitioned Markov Networks Song Liu The Inst. of Stats. Math., Taiji Suzuki , Masashi Sugiyama University of Tokyo, Kenji Fukumizu The Institute of Statistical MathematicsPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • Best paper award – 05:44 – Ensuring Rapid Mixing and Low Bias for Asynchronous Gibbs Sampling Christopher De Sa Stanford, Chris ReStanford University, Kunle Olukotun StanfordPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 06:01 – Estimation from Indirect Supervision with Linear Moments Aditi Raghunathan Stanford University, Roy Frostig Stanford University, John , Percy Liang StanfordPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    ⤴

    Tuesday – Supervised Learning

    Session chair: Nicolas Le Roux

    Location:  Liberty
  • 05:10 – Early and Reliable Event Detection Using Proximity Space Representation Maxime Sangnier LTCI, CNRS, Télécom ParisTech, Jerome Gauthier CEA, Alain RakotomamonjyPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 05:27 – Meta–Gradient Boosted Decision Tree Model for Weight and Target Learning Yury Ustinovskiy Yandex, Valentina FedorovaYandex, Gleb Gusev Yandex, Pavel Serdyukov YandexPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 05:44 – Differential Geometric Regularization for Supervised Learning of Classifiers Qinxun Bai Boston University, Steven RosenbergBoston University, Zheng Wu The Mathwork Inc., Stan Sclaroff Boston UniversityPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 06:01 – Linking losses for density ratio and class-probability estimation Aditya Menon National ICT Australia, Cheng Soon Ong NICTAPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    ⤴

    Wednesday – Neural Networks and Deep Learning I

    Session chair: Samy Bengio

    Location:  Ballroom 1+2+Juliard
  • 08:30 – Neural Variational Inference for Text Processing Yishu Miao University of Oxford, Lei Yu University of Oxford, Phil BlunsomPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 08:47 – A Deep Learning Approach to Unsupervised Ensemble Learning Uri Shaham Yale University, Xiuyuan Cheng , Omer Dror , Ariel Jaffe , Boaz Nadler , Joseph Chang , Yuval KlugerPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:04 – From Softmax to Sparsemax: A Sparse Model of Attention and Multi-Label Classification Andre Martins , Ramon Astudillo UnbabelPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:21 – A Neural Autoregressive Approach to Collaborative Filtering Yin Zheng Hulu LLC., Bangsheng Tang Hulu LLC, Wenkui Ding Hulu LLC., Hanning Zhou HuluPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:38 – Scalable Gradient-Based Tuning of Continuous Regularization Hyperparameters Jelena Luketina Aalto University, Tapani Raiko , Mathias Berglund Aalto University, Klaus Greff IDSIAPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract

    ⤴

    Wednesday – Optimization (Continuous)

    Session chair: Shie Mannor

    Location:  Ballroom 3+4
  • 08:30 – SDNA: Stochastic Dual Newton Ascent for Empirical Risk Minimization Zheng Qu University of Hong Kong, Peter Richtarik , Martin Takac , Olivier Fercoq LTCI, CNRS, Télécom ParisTech, Université Paris-Saclay, 75013, Paris, FrancePaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 08:47 – Stochastic Block BFGS: Squeezing More Curvature out of Data Robert Gower University of Edinburgh, Donald Goldfarb Columbia University, Peter RichtarikPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:04 – PD-Sparse : A Primal and Dual Sparse Approach to Extreme Multiclass and Multilabel Classification Ian En-Hsu Yen University of Texas at Austin, Xiangru Huang UTaustin, Pradeep Ravikumar UT Austin, Kai Zhong ICES department, University of Texas at Austin, InderjitPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:21 – Parallel and Distributed Block-Coordinate Frank-Wolfe Algorithms Yu-Xiang Wang , Veeranjaneyulu Sadhanala , Wei Dai Carnegie Mellon University, Willie Neiswanger , Suvrit Sra , Eric Xing CMUPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:38 – Minding the Gaps for Block Frank-Wolfe Optimization of Structured SVMs Anton Osokin , Jean-Baptiste Alayrac ENS, Isabella Lukasewitz INRIA, Puneet Dokania INRIA and Ecole Centrale Paris, Simon Lacoste-Julien INRIAPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract

    Session chair: Satyen Kale

    Location:  Ballroom 3+4
  • 10:20 – On the Iteration Complexity of Oblivious First-Order Optimization Algorithms Yossi Arjevani Weizmann Institute of Science, Ohad Shamir Weizmann Institute of SciencePaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 10:37 – Variance-Reduced and Projection-Free Stochastic Optimization Elad Hazan Princeton University, Haipeng LuoPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 10:54 – On Graduated Optimization for Stochastic Non-Convex Problems Elad Hazan Princeton University, Kfir Yehuda Levy Technion, Shai Shalev-Shwartz Hebrew University of JerusalemPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 11:30 – A Self-Correcting Variable-Metric Algorithm for Stochastic Optimization Frank Curtis Lehigh UniversityPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 11:47 – A Superlinearly-Convergent Proximal Newton-type Method for the Optimization of Finite Sums Anton Rodomanov Higher School of Economics, Dmitry Kropotov Moscow State UniversityPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 12:04 – Stochastic Variance Reduced Optimization for Nonconvex Sparse Learning Xingguo Li University of Minnesota, Tuo Zhao , Raman Arora Johns Hopkins University, Han , Jarvis HauptPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 03:40 – Energetic Natural Gradient Descent Philip Thomas CMU, Bruno Castro da Silva , Christoph Dann Carnegie Mellon University, EmmaPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 03:57 – On the Quality of the Initial Basin in Overspecified Neural Networks Itay Safran Weizmann Institute of Science, Ohad ShamirWeizmann Institute of SciencePaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:14 – L1-regularized Neural Networks are Improperly Learnable in Polynomial Time Yuchen Zhang , Jason , MichaelPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    ⤴

    Wednesday – Multi-label, multi-task, and neural networks

    Session chair: Joelle Pineau

    Location:  Marquis
  • 08:30 – Asymmetric Multi-task Learning based on Task Relatedness and Confidence Giwoong Lee UNIST, Eunho Yang IBM Research, Sung ju Hwang UNISTPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 08:47 – Training Deep Neural Networks via Direct Loss Minimization Yang Song Tsinghua University, Alexander Schwing , Richard , Raquel Urtasun U. of TorontoPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:04 – Structured Prediction Energy Networks David Belanger University of Massachusetts Am, Andrew McCallumPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:21 – Conditional Bernoulli Mixtures for Multi-label Classification Cheng Li Northeastern University, Bingyu Wang Northeastern University, Virgil Pavlu Northeastern University, Javed Aslam Northeastern UniversityPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:38 – Training Neural Networks Without Gradients: A Scalable ADMM Approach Gavin Taylor US Naval Academy, Ryan Burmeister US Naval Academy, Zheng Xu University of Maryland, Bharat Singh University of Maryland, Colleg, Ankit Patel Rice University, Tom GoldsteinUniversity of MarylandPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract

    ⤴

    Wednesday – Gaussian Processes

    Session chair: Roger Grosse

    Location:  Lyceum
  • 08:30 – Stability of Controllers for Gaussian Process Forward Models Julia Vinogradska Robert Bosch GmBH, Bastian Bischoff , Duy Nguyen-Tuong , Anne Romer , Henner Schmidt , Jan PetersPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 08:47 – A Distributed Variational Inference Framework for Unifying Parallel Sparse Gaussian Process Regression Models Trong Nghia Hoang NUS, Quang Minh Hoang NUS, Bryan Kian Hsiang Low NUSPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:04 – Deep Gaussian Processes for Regression using Approximate Expectation Propagation Thang Bui University of Cambridge, Daniel Hernandez-Lobato , Jose miguel Hernandez-Lobato , Yingzhen Li University of Cambridge, Richard Turner University of CambridgePaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:21 – Preconditioning Kernel Matrices Kurt Cutajar EURECOM, Michael Osborne , John Cunningham Columbia University, Maurizio Filippone EURECOMPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:38 – Extended and Unscented Kitchen Sinks Edwin Bonilla UNSW, Daniel Steinberg DATA61, Alistair Reid Data61Paper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract

    ⤴

    Wednesday – Feature Selection and Dimensionality Reduction

    Session chair: Laurens van der Maaten

    Location:  Empire
  • 08:30 – On the Consistency of Feature Selection With Lasso for Non-linear Targets Yue Zhang Case Western Reserve University, Weihong Guo Case Western Reserve University, Soumya RayPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 08:47 – No penalty no tears: Least squares in high-dimensional linear models Xiangyu Wang Duke University, David , Chenlei LengUniversity of WarwickPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:04 – Simultaneous Safe Screening of Features and Samples in Doubly Sparse Modeling Atsushi Shibagaki Nagoya Institute of Technology, Masayuki Karasuyama Nagoya Institute of Technology, Kohei Hatano Kyushu University, Ichiro TakeuchiPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:21 – Efficient Learning with a Family of Nonconvex Regularizers by Redistributing Nonconvexity Quanming Yao HKUST, James KwokHong Kong University Science TechnologyPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:38 – How to Fake Multiply by a Gaussian Matrix Michael Kapralov , Vamsi Potluru Comcast Cable, David WoodruffPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract

    ⤴

    Wednesday – Graph Analysis/ Spectral Methods

    Session chair: Alex Kulesza

    Location:  Soho
  • 08:30 – Metadata-conscious anonymous messaging Giulia Fanti UIUC, Peter Kairouz UIUC, Sewoong Oh UIUC, Kannan Ramchandran UC Berkeley, Pramod Viswanath UIUCPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 08:47 – A Simple and Strongly-Local Flow-Based Method for Cut Improvement Nate Veldt Purdue Unversity, David Gleich Purdue University, MichaelPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:04 – Community Recovery in Graphs with Locality Yuxin Chen Stanford University, Govinda Kamath Stanford University, Changho SuhKAIST, David Tse Stanford UniversityPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:21 – Interactive Bayesian Hierarchical Clustering Sharad Vikram UCSD, Sanjoy Dasgupta UCSDPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:38 – Cross-Graph Learning of Multi-Relational Associations Hanxiao Liu Carnegie Mellon University, Yiming YangPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract

    ⤴

    Wednesday – Ranking and Preference Learning

    Session chair: Tong Zhang

    Location:  Liberty
  • 08:30 – Controlling the distance to a Kemeny consensus without computing it Anna Korba Telecom Paris Tech, Yunlong Jiao Mines Paris Tech, Eric Sibony Telecom Paris TechPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 08:47 – Data-driven Rank Breaking for Efficient Rank Aggregation Ashish Khetan UIUC, Sewoong Oh UIUCPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:04 – Parameter Estimation for Generalized Thurstone Choice Models Milan Vojnovic Microsoft, Seyoung Yun MicrosoftPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:21 – Learning Mixtures of Plackett-Luce Models Zhibing Zhao RPI, Peter Piech RPI, Lirong Xia RPIPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 09:38 – Recommendations as Treatments: Debiasing Learning and Evaluation Tobias Schnabel Cornell University, Adith SwaminathanCornell University, Ashudeep Singh Cornell University, Navin Chandak Google, Thorsten Joachims CornellPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract

    ⤴

    Wednesday – Neural Networks and Deep Learning

    Session chair: Yoshua Bengio

    Location:  Ballroom 1+2+Juliard
  • 10:20 – Generative Adversarial Text to Image Synthesis Scott Reed , Zeynep Akata Max Planck Institute for Informatics, Xinchen YanUniversity of Michigan, Lajanugen Logeswaran University of Michigan – Ann Arbor, Bernt Schiele , Honglak Lee University of MichiganPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 10:37 – Autoencoding beyond pixels using a learned similarity metric Anders Boesen Lindbo Larsen Technical University of Denmar, Søren Kaae Sønderby University of Copenagen, Hugo Larochelle Twitter, Ole Winther Technical University of DenmarkPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 10:54 – Exploiting Cyclic Symmetry in Convolutional Neural Networks Sander Dieleman Google DeepMind, Jeffrey De Fauw Google DeepMind, Koray Kavukcuoglu Google DeepmindPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 11:30 – A Comparative Analysis and Study of Multiview CNN Models for Joint Object Categorization and Pose Estimation Mohamed Elhoseiny Rutgers University, Tarek El-Gaaly Rutgers University, Amr Bakry Rutgers University, Ahmed Elgammal Rutgers UniversityPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 11:47 – Dynamic Capacity Networks Amjad Almahairi , Nicolas Ballas , Tim Cooijmans University of Montreal, Yin Zheng Hulu LLC., Hugo Larochelle Twitter, Aaron CourvillePaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 12:04 – Augmenting Supervised Neural Networks with Unsupervised Objectives for Large-scale Image Classification Yuting ZhangUniversity of Michigan, Kibok Lee University of Michigan, Honglak Lee University of MichiganPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract

    ⤴

    Wednesday – Applications and Time-Series Analysis

    Session chair: Jon Mcauliffe

    Location:  Marquis
  • 10:20 – Hierarchical Decision Making In Electricity Grid Management Gal Dalal Technion, Elad Gilboa Technion, Shie Mannor TechnionPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 10:37 – ForecastICU: A Prognostic Decision Support System for Timely Prediction of Intensive Care Unit Admission Jinsung YoonUniversity of California, Los , Ahmed Alaa University of California, Los Angeles, Scott Hu University of California, Los Angeles, Mihaela van der SchaarPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 10:54 – Power of Ordered Hypothesis Testing Lihua Lei Lihua, William Fithian UC Berkeley, Department of StatisticsPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 11:30 – Learning to Filter with Predictive State Inference Machines Wen Sun Carnegie Mellon University, Arun Venkatraman Carnegie Mellon University, Byron Boots , J.Andrew Bagnell Carnegie Mellon UniversityPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 11:47 – Learning Population-Level Diffusions with Generative RNNs Tatsunori Hashimoto MIT, David Gifford MIT, Tommi Jaakkola MITPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 12:04 – Fast Parameter Inference in Nonlinear Dynamical Systems using Iterative Gradient Matching Mu Niu University of Glasgow, Simon Rogers University of Glasgow, Maurizio Filippone EURECOM, Dirk Husmeier University of GlasgowPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract

    ⤴

    Wednesday – Dimensionality Reduction / Private Learning

    Session chair: Jeff Bilmes

    Location:  Lyceum
  • 10:20 – Greedy Column Subset Selection: New Bounds and Distributed Algorithms Jason Altschuler Princeton University, Aditya Bhaskara, Gang Fu Google Research, Vahab Mirrokni Google Research, Afshin Rostamizadeh Google, Morteza Zadimoghaddam Google ResearchPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 10:37 – Efficient Private Empirical Risk Minimization for High-dimensional Learning Shiva Prasad Kasiviswanathan Samsung Research America, Hongxia Jin Samsung Research AmericaPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 10:54 – Binary embeddings with structured hashed projections Anna Choromanska Courant Institute, NYU, Krzysztof Choromanski Google Research NYC, Mariusz Bojarski NVIDIA, Tony Jebara Columbia, Sanjiv Kumar , YannPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 11:30 – Differentially Private Policy Evaluation Borja Balle Lancaster University, Maziar Gomrokchi McGill University, Doina Precup McGillPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 11:47 – Learning from Multiway Data: Simple and Efficient Tensor Regression Rose Yu University of Southern Cal, Yan LiuPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 12:04 – Low-Rank Matrix Approximation with Stability Dongsheng Li IBM Research – China, Chao Chen , Qin Lv , Junchi Yan , Li Shang , Stephen ChuPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract

    ⤴

    Wednesday – Monte Carlo Methods

    Session chair: Stefano Ermon

    Location:  Empire
  • 10:20 – Interacting Particle Markov Chain Monte Carlo Tom Rainforth University of Oxford, Christian Naesseth Linköping University, Fredrik Lindsten Uppsala University, Brooks Paige University of Oxford, Jan-Willem Vandemeent , Arnaud Doucet University of Oxford, Frank WoodPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 10:37 – Slice Sampling on Hamiltonian Trajectories Benjamin Bloem-Reddy Columbia University, John Cunningham Columbia UniversityPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 10:54 – Robust Monte Carlo Sampling using Riemannian Nos\'{e}-Poincar\'{e} Hamiltonian Dynamics Anirban Roychowdhury , Brian KulisBoston University, Srinivasan ParthasarathyPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 11:30 – Inference Networks for Sequential Monte Carlo in Graphical Models Brooks Paige University of Oxford, Frank WoodPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 11:47 – Partition Functions from Rao-Blackwellized Tempered Sampling David Carlson Columbia University, Patrick Stinson Columbia University, Ari Pakman Columbia University, LiamPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 12:04 – Stochastic Quasi-Newton Langevin Monte Carlo Umut Simsekli Telecom ParisTech, Roland Badeau , Taylan Cemgil , Gaël RichardPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract

    ⤴

    Wednesday – Crowdsourcing and Interactive Learning

    Session chair: Yisong Yue

    Location:  Soho
  • 10:20 – No Oops, You Won’t Do It Again: Mechanisms for Self-correction in Crowdsourcing Nihar Shah UC Berkeley, Dengyong ZhouMicrosoft ResearchPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 10:37 – The Label Complexity of Mixed-Initiative Classifier Training Jina Suh Microsoft, Xiaojin Zhu University of Wisconsin, Saleema Amershi MicrosoftPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 10:54 – The Knowledge Gradient for Sequential Decision Making with Stochastic Binary Feedbacks Yingfei Wang Princeton University, Chu Wang , Warren PowellPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 11:30 – Estimating Accuracy from Unlabeled Data: A Bayesian Approach Emmanouil Antonios Platanios Carnegie Mellon University, Avinava Dubey Carnegie Mellon University, Tom Mitchell Carnegie Mellon UniversityPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 11:47 – Actively Learning Hemimetrics with Applications to Eliciting User Preferences Adish Singla , Sebastian Tschiatschek ETH Zurich, Andreas KrausePaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 12:04 – Optimality of Belief Propagation for Crowdsourced Classification Jungseul Ok KAIST, Sewoong Oh UIUC, Jinwoo Shin KAIST, Yung Yi KAISTPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract

    ⤴

    Wednesday – Learning Theory

    Session chair: Alina Beygelzimer

    Location:  Liberty
  • 10:20 – Improved SVRG for Non-Strongly-Convex or Sum-of-Non-Convex Objectives Zeyuan Allen-Zhu Princeton University, Yang YuanCornellPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 10:37 – Variance Reduction for Faster Non-Convex Optimization Zeyuan Allen-Zhu Princeton University, Elad Hazan Princeton UniversityPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 10:54 – Even Faster Accelerated Coordinate Descent Using Non-Uniform Sampling Zeyuan Allen-Zhu Princeton University, Zheng Qu The University of Hong Kong, Peter Richtarik , Yang Yuan CornellPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 11:30 – False Discovery Rate Control and Statistical Quality Assessment of Annotators in Crowdsourced Ranking QianQian Xu IIE, CAS, Jiechao Xiong Peking University, Xiaochun Cao Institute of information engineering, CAS, Yuan Yao Peking UniversityPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 11:47 – On the Power and Limits of Distance-Based Learning Periklis Papakonstantinou Rutgers University , Jia Xu Chinese Academy of Sciences, Guang Yang ICT, BeijingPaper | Reviews | Rebuttal | Poster session on tuesday afternoon 3pm-7pm | Abstract
  • 12:04 – Train faster, generalize better: Stability of stochastic gradient descent Moritz Hardt Google, Ben Recht Berkeley, YoramPaper | Reviews | Rebuttal | Poster session on tuesday morning 10am-1pm | Abstract

    ⤴

    Wednesday – Supervised Learning

    Session chair: Csaba Szepesvari

    Location:  Ballroom 3+4
  • 03:40 – Sparse Nonlinear Regression: Parameter Estimation Zhuoran Yang Princeton University, Zhaoran Wang Princeton University, Han , Yonina Eldar Technion, Tong ZhangPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 03:57 – Polynomial Networks and Factorization Machines: New Insights and Efficient Training Algorithms Mathieu Blondel NTT, Masakazu Ishihata NTT Communication Science Labo, Akinori Fujino NTT, Naonori UedaPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:14 – Hyperparameter optimization with approximate gradient Fabian Pedregosa INRIAPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    ⤴

    Wednesday – Kernel Methods

    Session chair: Marius Kloft

    Location:  Marquis
  • 03:40 – DR-ABC: Approximate Bayesian Computation with Kernel-Based Distribution Regression Jovana Mitrovic University of Oxford, Dino Sejdinovic University of Oxford, Yee-Whye Teh OxfordPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 03:57 – Persistence weighted Gaussian kernel for topological data analysis Genki Kusano Tohoku University, Yasuaki Hiraoka , Kenji Fukumizu The Institute of Statistical MathematicsPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:14 – Discriminative Embeddings of Latent Variable Models for Structured Data Hanjun Dai Georgia Tech, Bo Dai Georgia Tech, Le SongGatechPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    ⤴

    Wednesday – Matrix Factorization and Related Topics

    Session chair: Stefanie Jegelka

    Location:  Lyceum
  • 03:40 – Recycling Randomness with Structure for Sublinear time Kernel Expansions Krzysztof Choromanski Google Research NYC, Vikas Sindhwani Google ResearchPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 03:57 – Optimal Classification with Multivariate Losses Nagarajan Natarajan Microsoft Research India, Oluwasanmi Koyejo Stanford University & University of Illinois at Urbana Champaign, Pradeep Ravikumar UT Austin, InderjitPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:14 – Sparse Parameter Recovery from Aggregated Data Avradeep Bhowmik University of Texas at Austin, Joydeep , Oluwasanmi KoyejoStanford University & University of Illinois at Urbana ChampaignPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    ⤴

    Wednesday – Privacy, Anonymity, and Security

    Session chair: Gal Chechik

    Location:  Empire
  • 03:40 – Learning privately from multiparty data Jihun Hamm , Yingjun Cao UC-San Diego, Mikhail BelkinPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 03:57 – Differentially Private Chi-Squared Hypothesis Testing: Goodness of Fit and Independence Testing Ryan Rogers University of Pennsylvania, Salil Vadhan Harvard University, Hyun Lim UCLA, Marco Gaboardi University at BuffaloPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:14 – Discrete Distribution Estimation under Local Privacy Peter Kairouz UIUC, Keith Bonawitz Google, Daniel RamagePaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    ⤴

    Wednesday – Causal Inference

    Session chair: David Sontag

    Location:  Soho
  • 03:40 – The Arrow of Time in Multivariate Time Series Stefan Bauer ETH Zurich, Bernhard Schölkopf , JonasPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 03:57 – Conditional Dependence via Shannon Capacity: Axioms, Estimators and Applications Weihao Gao UIUC, Sreeram Kannan UW Seattle, Sewoong Oh UIUC, Pramod Viswanath UIUCPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:14 – Learning Granger Causality for Hawkes Processes Hongteng Xu Georgia Tech, Mehrdad Farajtabar , HongyuanPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    ⤴

    Wednesday – Optimization

    Session chair: Lihong Li

    Location:  Liberty
  • 03:40 – Gossip Dual Averaging for Decentralized Optimization of Pairwise Functions Igor Colin , Aurelien Bellet INRIA, Joseph Salmon , Stéphan ClémençonPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 03:57 – Adaptive Sampling for SGD by Exploiting Side Information Siddharth GopalPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract
  • 04:14 – Mixture Proportion Estimation via Kernel Embeddings of Distributions Harish Ramaswamy , Clayton , AmbujPaper | Reviews | Rebuttal | Poster session on wednesday morning 10am-1pm | Abstract

    • 你可能感兴趣的:(机器学习,深度学习)

    • 图像识别与应用 狂踹瘸子那条好脚 python
      图像识别作为人工智能领域的重要分支，近年来取得了显著进展，其中卷积神经网络（CNN）功不可没。CNN凭借其强大的特征提取能力，在图像分类、目标检测、人脸识别等任务中表现出色，成为图像识别领域的核心技术。一、卷积神经网络：图像识别的利器CNN是一种专门处理网格状数据的深度学习模型，其结构设计灵感来源于生物视觉系统。与全连接神经网络不同，CNN通过卷积层、池化层等结构，能够有效提取图像的局部特征，并逐
    • 大模型如何改变教育？典型应用场景的探究与展望！ AGI大模型学习 大模型应用人工智能AI产品经理llama大模型AI大模型教程
      目前，大模型在教育领域的应用主要体现在个性化学习助手、智能问答系统、内容生成与创作辅助、智能写作评估、跨语言学习支持、数学解题辅助等几个方面。大模型技术在教育领域凭借卓越的数据处理能力和深度学习技术，极大推动了教育质量的提升与教育公平的实现。分级分类的教育数据助力大模型发展在构建与优化大模型的过程中，教育数据能够帮助我们更精准地理解教育现象，更有质量地辅助教学。教育数据涵盖广泛，包括但不限于学生的
    • Python中的 redis keyspace 通知_python 操作redis psubscribe(‘__keyspace@0__ ‘) 2301_82243733 程序员python学习面试
      最后Python崛起并且风靡，因为优点多、应用领域广、被大牛们认可。学习Python门槛很低，但它的晋级路线很多，通过它你能进入机器学习、数据挖掘、大数据，CS等更加高级的领域。Python可以做网络应用，可以做科学计算，数据分析，可以做网络爬虫，可以做机器学习、自然语言处理、可以写游戏、可以做桌面应用…Python可以做的很多，你需要学好基础，再选择明确的方向。这里给大家分享一份全套的Pytho
    • DeepSeek原理介绍以及对网络安全行业的影响 AI拉呱 Deepseek人工智能
      大家好，我是AI拉呱，一个专注于人工智领域与网络安全方面的博主，现任资深算法研究员一职，兼职硕士研究生导师；热爱机器学习和深度学习算法应用，深耕大语言模型微调、量化、私域部署。曾获多次获得AI竞赛大奖，拥有多项发明专利和学术论文。对于AI算法有自己独特见解和经验。曾辅导十几位非计算机学生转行到算法岗位就业。关注评审分享一起学习更多知识。1.DeepSeek公司介绍1.1DeepSeek是什么：wh
    • auto-gptq安装以及不适配软硬件环境可能出现的问题及解决方式 IT修炼家 大模型部署大模型auto-gptqcuda
      目录1、auto-gptq是什么？2、auto-gptq安装3、auto-gptq不正确安装可能会出现的问题（1）爆出：`CUDAextensionnotinstalled.`（2）没有报错但是推理速度超级慢1、auto-gptq是什么？Auto-GPTQ是一种专注于量化深度学习模型的工具库。它的主要目标是通过量化技术（Quantization）将大型语言模型（LLM）等深度学习模型的大小和计算复
    • 【deepseek与chatGPT辩论】辩论题： “人工智能是否应当具备自主决策能力？” 海宁不掉头发 软件工程人工智能人工智能chatgptdeepseek
      探讨辩论题这个提案涉及创建一个精确的辩论题目，旨在测试deepseek的应答能力。创建辩论题目提议设计一个辩论题目以测试deepseek的应答能力。希望这个题目具有挑战性并能够测量其回应质量。好的，来一道适合深度学习的辩论题：辩论题：“人工智能是否应当具备自主决策能力？”这个话题涉及到人工智能的发展、伦理以及未来应用，可以从以下几个方面展开辩论：支持方：认为人工智能的自主决策能力能够加速科技进步，
    • FOKS-TROT: 一个高效、易用的全功能开源知识图谱生成工具 柳旖岭
      FOKS-TROT:一个高效、易用的全功能开源知识图谱生成工具项目简介FOKS-TROT是一个基于Python的全功能开源知识图谱生成工具，旨在帮助研究人员和开发者快速构建具有丰富信息的知识图谱。该项目由hkx3upper在GitCode上开发并维护。通过FOKS-TROT，您可以轻松地将各种数据源（如文本文件、数据库、API）转换为结构化的知识图谱，并对其进行可视化分析和机器学习任务。此外，该工
    • 基于python深度学习遥感影像地物分类与目标识别、分割实践技术应用 xiao5kou4chang6kai4 深度学习遥感勘测python深度学习分类
      专题一：深度学习发展与机器学习深度学习的历史发展过程机器学习，深度学习等任务的基本处理流程梯度下降算法讲解不同初始化，学习率对梯度下降算法的实例分析从机器学习到深度学习算法专题二深度卷积网络、卷积神经网络、卷积运算的基本原理池化操作，全连接层，以及分类器的作用BP反向传播算法的理解一个简单CNN模型代码理解特征图，卷积核可视化分析专题三TensorFlow与keras介绍与入门TensorFlow
    • LLM与知识图谱融合:智能运维知识库构建 AI天才研究院 DeepSeekR1&大数据AI人工智能大模型AI大模型企业级应用开发实战AI实战计算科学神经计算深度学习神经网络大数据人工智能大型语言模型AIAGILLMJavaPython架构设计AgentRPA
      1.背景介绍随着信息技术的飞速发展，IT运维管理面临着越来越大的挑战。海量的设备、复杂的网络环境、日益增长的数据量，使得传统的运维方式难以满足需求。为了提高运维效率和质量，智能运维应运而生。智能运维的核心是将人工智能技术应用于运维领域，通过机器学习、深度学习等算法，实现自动化、智能化的运维管理。其中，大语言模型（LLM）和知识图谱是两个重要的技术方向。LLM能够理解和生成自然语言，可以用于构建智能
    • 机器学习·文本数据读写处理 AAA顶置摸鱼 python深度学习机器学习人工智能数据处理
      前言在自然语言处理的第一步，需要面对的是各种各样以不同形式表现的文本数据，比如，txt、Excel中的表格数据，还有无法直接打开的pkl文件等。针对这些不同类型的数据，可以基于Python中的基本功能函数或者调用某些库进行读写以及作一些基本的处理。一、文本数据读写方法1.读写TXT文件读取方法：read()：读取整个文件，返回字符串。readline()：逐行读取，返回字符串。readlines(
    • 深度应用场景：DeepSeek —— 探索AI赋能的智慧未来 人工智能专属驿站 人工智能
      深度应用场景：DeepSeek——探索AI赋能的智慧未来随着人工智能的迅猛发展，数据的价值已不再局限于简单的存储与处理，它们正变得更加智能与高效。DeepSeek，这一创新的AI技术平台，正以其独特的深度学习能力，开启了各行各业的智能化变革。让我们走进一个由DeepSeek打造的深度应用场景，探索它如何推动未来的发展。1.智能医疗：精准诊断，拯救生命想象一下，医生们不再是唯一的诊断专家，而是与AI
    • 卷积神经网络之AlexNet经典神经网络，实现手写数字0~9识别 知识鱼丸 深度学习神经网络cnn人工智能深度学习AlexNet经典神经网络
      深度学习中较为常见的神经网络模型AlexNet，AlexNet是一个采用GPU训练的深层CNN，本质是种LeNet变体。由特征提取层的5个卷积层两个下采样层和分类器中的三个全连接层构成。先看原理：AlexNet网络特点采用ReLU激活函数，使训练速度提升6倍采用dropout层，防止模型过拟合通过平移和翻转的方式对数据进行增强采用LRN局部响应归一化，限制数据大小，防止梯度消失和爆炸。但后续证明批
    • 用 TensorFlow 搭建简单的手写数字识别模型 lozhyf 工作面试学习tensorflow人工智能python
      一、引言手写数字识别是机器学习领域中一个经典且基础的问题，它在很多实际场景中都有广泛的应用，比如邮政系统中的邮件分拣、银行支票金额识别等。TensorFlow是一个强大的开源机器学习框架，由Google开发并维护，它提供了丰富的工具和接口，能帮助我们快速搭建和训练深度学习模型。在这篇博客中，我们将使用TensorFlow构建一个简单的神经网络模型，用于识别手写数字。二、环境准备在开始之前，你需要安
    • 深度学习在医疗影像分析中的革命性应用 Echo_Wish 人工智能前沿技术深度学习人工智能
      深度学习在医疗影像分析中的革命性应用引言医疗影像分析是现代医学中不可或缺的一部分，特别是在疾病诊断和治疗过程中发挥了至关重要的作用。随着深度学习技术的发展，医疗影像分析的效率和准确性得到了显著提升。本文将探讨如何利用深度学习技术，特别是Python编程语言，来优化医疗影像分析，展示具体的代码实例，并举例说明其实际应用效果。深度学习与医疗影像分析深度学习（DeepLearning）是一种基于人工神经
    • yolov8人脸识别与脸部关键点检测（代码+原理） QQ_1309399183 计算机视觉实战项目集锦YOLO人工智能人脸识别yolo人脸检测
      YOLOv8脸部识别是一个基于YOLOv8算法的人脸检测项目，旨在实现快速、准确地检测图像和视频中的人脸。该项目是对YOLOv8算法的扩展和优化，专门用于人脸检测任务。YOLOv8是一种基于深度学习的目标检测算法，通过将目标检测问题转化为一个回归问题，可以实现实时的目标检测。YOLOv8Face项目在YOLOv8的基础上进行了改进，使其更加适用于人脸检测。以下是YOLOv8Face项目的一些特点和
    • 【机器学习】基于3D CNN通过CT图像分类预测肺炎 MUKAMO AIPython应用机器学习深度学习人工智能神经网络3DCNN
      1.引言1.1.研究背景在医学诊断中，医生通过分析CT影像来预测疾病时，面临一些挑战和局限性：图像信息的广度与复杂性：CT扫描生成的大量图像对医生来说既是信息的宝库也是处理上的负担。每组CT数据可能包含数百张切片，医生必须迅速审阅这些图像，以便捕捉到病变的微小细节。这种庞大的信息量要求医生在有限的时间内做出精准诊断，但同时也增加了漏诊或误诊的风险。部分容积效应也可能模糊小病变的边界，使得准确诊断变
    • TensorFlow LiteRT 概览 姚家湾 tensorflow人工智能python
      LiteRT（简称LiteRuntime，以前称为TensorFlowLite）是Google面向设备端AI的高性能运行时。您可以找到适用于各种机器学习/AI任务的LiteRT就绪模型，也可以使用AIEdge转换和优化工具将TensorFlow、PyTorch和JAX模型转换为TFLite格式并运行。主要特性针对设备端机器学习进行了优化：LiteRT解决了五项关键的ODML约束条件：延迟时间（无需
    • 机器学习（1）安装Pytorch CoderIsArt 机器学习与深度学习机器学习pytorch人工智能
      1.安装命令pip3installtorchtorchvisiontorchaudio--index-urlhttps://download.pytorch.org/whl/cu1182.安装过程Log：Lookinginindexes:https://download.pytorch.org/whl/cu118CollectingtorchDownloadinghttps://download.
    • 探索Vearch：高效的深度学习向量相似度搜索系统 scaFHIO 深度学习人工智能python
      Vearch是一个可扩展的分布式系统，用于高效搜索深度学习向量的相似度。在本文中，我们将介绍Vearch的技术背景及其核心原理，演示如何使用VearchPythonSDK进行安装和设置，并分析一些实际应用场景，最后提供一些实战建议。技术背景介绍随着深度学习技术的发展，向量相似度搜索在各类应用中变得越来越重要。从图像识别、推荐系统到自然语言处理，向量搜索可以极大地提升系统的性能。然而，随着数据量的增
    • 27岁大龄转码秋招惨败，朋友劝我转Java来得及吗？还是继续走前端或机器学习？ 程序员yt java机器学习开发语言
      今天给大家分享的是一位粉丝的提问，27岁大龄转码秋招惨败，朋友劝我转Java来得及吗？还是继续走前端或机器学习？接下来把粉丝的具体提问和我的回复分享给大家，希望也能给一些类似情况的小伙伴一些启发和帮助。同学提问：211建筑本科，22年毕业后gap一年转码去了英国读的QS100的it的水硕（24年12月份毕业），转码后对就业形势认知不足，时间全花在课业上，八股文和算法准备的不充足，秋招算是惨败。读研
    • 华为的云端训练算力与迭代效率 AI大模型应用之禅 DeepSeekR1&AI大模型与大数据计算科学神经计算深度学习神经网络大数据人工智能大型语言模型AIAGILLMJavaPython架构设计AgentRPA
      华为云、云端训练、算力、迭代效率、人工智能、深度学习、模型训练、分布式训练、优化算法1.背景介绍人工智能（AI）技术近年来发展迅速，深度学习作为其核心驱动力，在图像识别、自然语言处理、语音识别等领域取得了突破性进展。然而，深度学习模型的训练需要海量数据和强大的计算资源，这成为AI技术发展面临的瓶颈之一。云计算作为一种新型的计算模式，为深度学习提供了强大的算力支持。华为云作为国内领先的云计算平台，在
    • AI外呼机器人：营销新利器还是骚扰电话的升级版？ yoloGina 客户管理外呼系统电话外呼人工智能机器人
      "您好，这里是XX房产，最近有购房需求吗？""您好，您最近有种牙需求吗？"相信很多人都接到过类似的营销电话，而电话那头，很可能已经不是真人，而是AI外呼机器人。近年来，AI外呼系统凭借其高效率、低成本的优势，迅速在电销行业普及，成为企业营销的"新宠"。据统计，2022年中国AI外呼市场规模已达50亿元，预计2025年将突破100亿元。AI外呼系统的核心技术是语音识别和自然语言处理。通过深度学习海量
    • 深入浅出：CUDA是什么，如何利用它进行高效并行计算 码上飞扬 CUDA
      在当今这个数据驱动的时代，计算能力的需求日益增加，特别是在深度学习、科学计算和图像处理等领域。为了满足这些需求，NVIDIA推出了CUDA（ComputeUnifiedDeviceArchitecture），这是一种并行计算平台和编程模型。本文将带你全面了解CUDA的基本概念、工作原理及其应用场景。一、什么是CUDA？CUDA（ComputeUnifiedDeviceArchitecture）是由
    • 【深度学习】学习率调度策略 黑白交界 深度学习学习深度学习
      什么是学习率可以理解为模型在每一次迭代中的模型更新调整的幅度，“学习”新信息的速度。学习率定义了模型权重（参数）在梯度下降或其他优化算法中的更新步伐。较大的学习率意味着在每次参数更新时，模型会进行更大幅度的调整，而较小的学习率则意味着细致的、渐进的调整。适当的学习率可以帮助模型跳出局部最优解。当使用较大的学习率时，模型有可能跨越一些小的局部最优，从而找到全局最优解，但也有可能错过全局最优。因此，在
    • 【深度学习pytorch-93】Transformer 相比 RNN 的优势 华东算法王 DL-pytorch深度学习pytorchtransformer
      Transformer相比RNN的优势Transformer和RNN（循环神经网络）都是自然语言处理（NLP）领域的重要架构，但它们的工作原理和应用方式有很大不同。Transformer由于其独特的结构和机制，在多个方面优于RNN。以下是Transformer相比RNN的主要优势：1.并行计算能力RNN的局限性RNN是按顺序处理输入的，即每个时间步的输出都依赖于前一个时间步的输出。这意味着，在训练
    • 深度学习模型中的知识蒸馏是如何工作的? c++服务器开发 深度学习人工智能
      深度学习模型在多个领域，特别是计算机视觉和自然语言处理中，已经取得了革命性的进展。然而，随着模型复杂性和资源需求的不断攀升，如何将这些庞大模型的知识浓缩为更紧凑、更高效的形式，成为了当前研究的热点。知识蒸馏，作为一种将知识从复杂模型转移到更简单模型的策略，已经成为实现这一目标的有效工具。在本文中，我们将深入探究深度学习模型中知识蒸馏的概念、原理及其在各领域的应用，以期为读者提供一个全面而严谨的视角
    • 【核心算法篇七】《DeepSeek异常检测：孤立森林与AutoEncoder对比》 再见孙悟空_ 「2025DeepSeek技术全景实战」算法分布式docker计算机视觉人工智能自然语言处理DeepSeek
      大家好，今天我们来深入探讨一下《DeepSeek异常检测：孤立森林与AutoEncoder对比》这篇技术博客。我们将从核心内容、原理、应用场景等多个方面进行详细解析，力求让大家对这两种异常检测方法有一个全面而深入的理解。一、引言在数据科学和机器学习领域，异常检测（AnomalyDetection）是一个非常重要的任务。它的目标是从数据集中识别出那些与大多数数据显著不同的异常点。这些异常点可能是由于
    • 吐血整理！权重持久化方案优化，让你的模型性能飙升 盼达思文体科创 经验分享
      吐血整理！权重持久化方案优化，让你的模型性能飙升引言你是否在做深度学习项目时，遭遇过模型训练结果无法有效保存，导致之前的努力付诸东流的痛苦？又或者在模型权重持久化时，发现保存和加载的速度极慢，严重影响项目进度？今天咱们就来好好聊聊权重持久化方案的优化，帮你解决这些让人头疼的问题！核心内容❗传统方案痛点：大多数人都踩过的坑在很多深度学习项目里，大家常用的权重持久化方案存在不少问题。比如说，使用普通的
    • 吐血整理！模型热加载能力大比拼，谁才是真正王者？ 盼达思文体科创 经验分享
      吐血整理！模型热加载能力大比拼，谁才是真正王者？引言你是否在开发过程中，为了模型更新而频繁重启服务，浪费大量时间？又是否疑惑为什么有些模型加载速度快如闪电，而有些却慢得像蜗牛？今天就带你深入了解模型热加载能力的支持对比，让你不再为模型加载问题而烦恼！核心内容模型热加载概念科普场景化描述：想象一下，你正在运营一个基于机器学习模型的在线推荐系统。当你训练出了一个新的、性能更好的模型时，如果不能进行热加
    • Python从0到100（四）：Python中的运算符介绍(补充) 是Dream呀 pythonjava数据库
      前言：零基础学Python：Python从0到100最新最全教程。想做这件事情很久了，这次我更新了自己所写过的所有博客，汇集成了Python从0到100，共一百节课，帮助大家一个月时间里从零基础到学习Python基础语法、Python爬虫、Web开发、计算机视觉、机器学习、神经网络以及人工智能相关知识，成为学习学习和学业的先行者！欢迎大家订阅专栏：零基础学Python：Python从0到100最新
    • ztree异步加载 3213213333332132 JavaScriptAjaxjsonWebztree
      相信新手用ztree的时候,对异步加载会有些困惑，我开始的时候也是看了API花了些时间才搞定了异步加载，在这里分享给大家。 我后台代码生成的是json格式的数据，数据大家按各自的需求生成，这里只给出前端的代码。 设置setting，这里只关注async属性的配置 var setting = { //异步加载配置
    • thirft rpc 具体调用流程 BlueSkator 中间件rpcthrift
      Thrift调用过程中，Thrift客户端和服务器之间主要用到传输层类、协议层类和处理类三个主要的核心类，这三个类的相互协作共同完成rpc的整个调用过程。在调用过程中将按照以下顺序进行协同工作：         （1）     将客户端程序调用的函数名和参数传递给协议层（TProtocol），协议
    • 异或运算推导, 交换数据 dcj3sjt126com PHP异或^
      /* * 5 0101 * 9 1010 * * 5 ^ 5 * 0101 * 0101 * ----- * 0000 * 得出第一个规律: 相同的数进行异或, 结果是0 * * 9 ^ 5 ^ 6 * 1010 * 0101 * ---- * 1111 * * 1111 * 0110 * ---- * 1001
    • 事件源对象 周华华 JavaScript
      <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> <html xmlns="http://www.w3.org/1999/xhtml&q
    • MySql配置及相关命令 g21121 mysql
              MySQL安装完毕后我们需要对它进行一些设置及性能优化，主要包括字符集设置，启动设置，连接优化，表优化，分区优化等等。           一 修改MySQL密码及用户      
    • [简单]poi删除excel 2007超链接 53873039oycg Excel
            采用解析sheet.xml方式删除超链接，缺点是要打开文件2次,代码如下:      public void removeExcel2007AllHyperLink(String filePath) throws Exception { OPCPackage ocPkg = OPCPac
    • Struts2添加 open flash chart 云端月影
      准备以下开源项目： 1. Struts 2.1.6 2. Open Flash Chart 2 Version 2 Lug Wyrm Charmer (28th, July 2009) 3. jofc2，这东西不知道是没做好还是什么意思，好像和ofc2不怎么匹配，最好下源码，有什么问题直接改。 4. log4j 用eclipse新建动态网站，取名OFC2Demo，将Struts2 l
    • spring包详解 aijuans spring
      下载的spring包中文件及各种包众多，在项目中往往只有部分是我们必须的，如果不清楚什么时候需要什么包的话，看看下面就知道了。 aspectj目录下是在Spring框架下使用aspectj的源代码和测试程序文件。Aspectj是java最早的提供AOP的应用框架。 dist 目录下是Spring 的发布包，关于发布包下面会详细进行说明。 docs&nb
    • 网站推广之seo概念 antonyup_2006 算法Web应用服务器搜索引擎Google
         持续开发一年多的b2c网站终于在08年10月23日上线了。作为开发人员的我在修改bug的同时，准备了解下网站的推广分析策略。     所谓网站推广，目的在于让尽可能多的潜在用户了解并访问网站，通过网站获得有关产品和服务等信息，为最终形成购买决策提供支持。     网站推广策略有很多，seo，email，adv
    • 单例模式,sql注入,序列 百合不是茶 单例模式序列sql注入预编译
        序列在前面写过有关的博客,也有过总结,但是今天在做一个JDBC操作数据库的相关内容时 需要使用序列创建一个自增长的字段  居然不会了,所以将序列写在本篇的前面    1,序列是一个保存数据连续的增长的一种方式; 序列的创建; CREATE SEQUENCE seq_pro 2 INCREMENT BY 1 -- 每次加几个 3
    • Mockito单元测试实例 bijian1013 单元测试mockito
      Mockito单元测试实例： public class SettingServiceTest { private List<PersonDTO> personList = new ArrayList<PersonDTO>(); @InjectMocks private SettingPojoService settin
    • 精通Oracle10编程SQL(9)使用游标 bijian1013 oracle数据库plsql
      /* *使用游标 */ --显示游标 --在显式游标中使用FETCH...INTO语句 DECLARE CURSOR emp_cursor is select ename,sal from emp where deptno=1; v_ename emp.ename%TYPE; v_sal emp.sal%TYPE; begin ope
    • 【Java语言】动态代理 bit1129 java语言
        JDK接口动态代理 JDK自带的动态代理通过动态的根据接口生成字节码(实现接口的一个具体类)的方式，为接口的实现类提供代理。被代理的对象和代理对象通过InvocationHandler建立关联   package com.tom; import com.tom.model.User; import com.tom.service.IUserService;
    • Java通信之URL通信基础 白糖_ javajdkwebservice网络协议ITeye
      java对网络通信以及提供了比较全面的jdk支持，java.net包能让程序员直接在程序中实现网络通信。 在技术日新月异的现在，我们能通过很多方式实现数据通信，比如webservice、url通信、socket通信等等，今天简单介绍下URL通信。 学习准备：建议首先学习java的IO基础知识   URL是统一资源定位器的简写，URL可以访问Internet和www，可以通过url
    • 博弈Java讲义 - Java线程同步 (1) boyitech java多线程同步锁
        在并发编程中经常会碰到多个执行线程共享资源的问题。例如多个线程同时读写文件，共用数据库连接，全局的计数器等。如果不处理好多线程之间的同步问题很容易引起状态不一致或者其他的错误。    同步不仅可以阻止一个线程看到对象处于不一致的状态，它还可以保证进入同步方法或者块的每个线程，都看到由同一锁保护的之前所有的修改结果。处理同步的关键就是要正确的识别临界条件（cri
    • java-给定字符串，删除开始和结尾处的空格，并将中间的多个连续的空格合并成一个。 bylijinnan java
      public class DeleteExtraSpace { /** * 题目：给定字符串，删除开始和结尾处的空格，并将中间的多个连续的空格合并成一个。 * 方法1.用已有的String类的trim和replaceAll方法 * 方法2.全部用正则表达式，这个我不熟 * 方法3.“重新发明轮子”，从头遍历一次 */ public static v
    • An error has occurred.See the log file错误解决！ Kai_Ge MyEclipse
      今天早上打开MyEclipse时，自动关闭！弹出An error has occurred.See the log file错误提示！ 很郁闷昨天启动和关闭还好着！！！打开几次依然报此错误，确定不是眼花了！ 打开日志文件！找到当日错误文件内容： --------------------------------------------------------------------------
    • [矿业与工业]修建一个空间矿床开采站要多少钱? comsci
             地球上的钛金属矿藏已经接近枯竭...........        我们在冥王星的一颗卫星上面发现一些具有开采价值的矿床.....        那么,现在要编制一个预算,提交给财政部门..
    • 解析Google Map Routes dai_lm google api
      为了获得从A点到B点的路劲，经常会使用Google提供的API，例如 [url] http://maps.googleapis.com/maps/api/directions/json?origin=40.7144,-74.0060&destination=47.6063,-122.3204&sensor=false [/url] 从返回的结果上，大致可以了解应该怎么走，但
    • SQL还有多少“理所应当”？ datamachine sql
      转贴存档，原帖地址：http://blog.chinaunix.net/uid-29242841-id-3968998.html、http://blog.chinaunix.net/uid-29242841-id-3971046.html！ ------------------------------------华丽的分割线--------------------------------
    • Yii使用Ajax验证时，如何设置某些字段不需要验证 dcj3sjt126com Ajaxyii
      经常像你注册页面,你可能非常希望只需要Ajax去验证用户名和Email,而不需要使用Ajax再去验证密码,默认如果你使用Yii 内置的ajax验证Form,例如: $form=$this->beginWidget('CActiveForm', array(        'id'=>'usuario-form',&
    • 使用git同步网站代码 dcj3sjt126com crontabgit
      转自:http://ued.ctrip.com/blog/?p=3646?tn=gongxinjun.com   管理一网站，最开始使用的虚拟空间，采用提供商支持的ftp上传网站文件，后换用vps，vps可以自己搭建ftp的，但是懒得搞，直接使用scp传输文件到服务器，现在需要更新文件到服务器，使用scp真的很烦。发现本人就职的公司，采用的git+rsync的方式来管理、同步代码，遂
    • sql基本操作 蕃薯耀 sqlsql基本操作sql常用操作
      sql基本操作 >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> 蕃薯耀 2015年6月1日 17:30:33 星期一     &
    • Spring4+Hibernate4+Atomikos3.3多数据源事务管理 hanqunfeng Hibernate4
      Spring3+后不再对JTOM提供支持，所以可以改用Atomikos管理多数据源事务。Spring2.5+Hibernate3+JTOM参考：http://hanqunfeng.iteye.com/blog/1554251Atomikos官网网站：http://www.atomikos.com/   一.pom.xml <dependency> <
    • jquery中两个值得注意的方法one()和trigger()方法 jackyrong trigger
        在jquery中，有两个值得注意但容易忽视的方法，分别是one()方法和trigger()方法,这是从国内作者<<jquery权威指南》一书中看到不错的介绍 1） one方法     one方法的功能是让所选定的元素绑定一个仅触发一次的处理函数，格式为    one(type,${data},fn) &nb
    • 拿工资不仅仅是让你写代码的 lampcy 工作面试咨询
      这是我对团队每个新进员工说的第一件事情。这句话的意思是，我并不关心你是如何快速完成任务的，哪怕代码很差，只要它像救生艇通气门一样管用就行。这句话也是我最喜欢的座右铭之一。 这个说法其实很合理：我们的工作是思考客户提出的问题，然后制定解决方案。思考第一，代码第二，公司请我们的最终目的不是写代码，而是想出解决方案。 话粗理不粗。 付你薪水不是让你来思考的，也不是让你来写代码的，你的目的是交付产品
    • 架构师之对象操作----------对象的效率复制和判断是否全为空 nannan408 架构师
      1.前言。   如题。 2.代码。 (1)对象的复制，比spring的beanCopier在大并发下效率要高，利用net.sf.cglib.beans.BeanCopier Src src=new Src(); BeanCopier beanCopier = BeanCopier.create(Src.class, Des.class, false);
    • ajax 被缓存的解决方案 Rainbow702 JavaScriptjqueryAjaxcache缓存
      使用jquery的ajax来发送请求进行局部刷新画面，各位可能都做过。 今天碰到一个奇怪的现象，就是，同一个ajax请求，在chrome中，不论发送多少次，都可以发送至服务器端，而不会被缓存。但是，换成在IE下的时候，发现，同一个ajax请求，会发生被缓存的情况，只有第一次才会被发送至服务器端，之后的不会再被发送。郁闷。 解决方法如下： ① 直接使用 JQuery提供的 “cache”参数，
    • 修改date.toLocaleString()的警告 tntxia String
        我们在写程序的时候，经常要查看时间，所以我们经常会用到date.toLocaleString()，但是date.toLocaleString()是一个过时 的API，代替的方法如下：   package com.tntxia.htmlmaker.util; import java.text.SimpleDateFormat; import java.util.
    • 项目完成后的小总结 xiaomiya js总结项目
      项目完成了，突然想做个总结但是有点无从下手了。 做之前对于客户端给的接口很模式。然而定义好了格式要求就如此的愉快了。 先说说项目主要实现的功能吧 1，按键精灵 2，获取行情数据 3，各种input输入条件判断 4，发送数据（有json格式和string格式） 5，获取预警条件列表和预警结果列表， 6，排序， 7，预警结果分页获取 8，导出文件（excel，text等） 9，修
    按字母分类： ABCDEFGHIJKLMNOPQRSTUVWXYZ其他