今日焦点：卷积网络的问题及其解决方案CoordConv——CoordConv解决了坐标变换问题，具有更好的泛化能力，训练速度提高150倍，参数比卷积少10-100倍
《An Intriguing Failing of Convolutional Neural Networks and the CoordConv Solution》R Liu, J Lehman, P Molino... [Uber AI Labs] (2018) 网页链接 paper:网页链接 Pytorch implementation of CoordConv GitHub:网页链接
【Python 3教学课程(Jupyter notebooks)】’Jupyter notebooks for teaching/learning Python 3' by Jerry Pussinen GitHub: 网页链接
《Mask TextSpotter: An End-to-End Trainable Neural Network for Spotting Text with Arbitrary Shapes》P Lyu, M Liao, C Yao, W Wu, X Bai [Huazhong University of Science and Technology & Megvii (Face++) Technology] (2018) 网页链接 view:网页链接
《Multi-task Mid-level Feature Alignment Network for Unsupervised Cross-Dataset Person Re-Identification》S Lin, H Li, C Li, A C Kot [University of Warwick & Nanyang Technological University & Charles Sturt University] (2018) 网页链接 view:网页链接
【基于VORONOI图的“抢地盘”游戏】“Voronoi diagram area game” 网页链接
“(CMU)NeuLab Presentations at ACL 2018” 网页链接
【Learn how to design large-scale systems】网页链接 学习如何设计大规模系统--系统设计入门，为系统设计的面试做准备。被翻译成多种文字，中文翻译：网页链接 学习架构好资源，墙裂建议收藏！
“(colab)Build a linear model with Estimators” 网页链接
【ACL 2018 Facebook成果汇总】《Facebook Research at ACL 2018》 网页链接
《A network-based citation indicator of scientific performance》C Schulz, B Uzzi, D Helbing, O Woolley-Meza [ETH Zurich & Northwestern University] (2018) 网页链接 view:网页链接
《Scalable Gaussian Processes with Grid-Structured Eigenfunctions (GP-GRIEF)》T W. Evans, P B. Nair [University of Toronto] (2018) 网页链接 view:网页链接 GitHub:网页链接
《Combinatorial Bandits for Incentivizing Agents with Dynamic Preferences》T Fiez, S Sekar, L Zheng, L J. Ratliff [University of Washington] (2018) 网页链接 view:网页链接
《Using deep learning for comprehensive, personalized forecasting of Alzheimer's Disease progression》C K. Fisher, A M. Smith, J R. Walsh, t C A M Diseases [Unlearn.AI] (2018) 网页链接 view:网页链接
《Manifold regularization with GANs for semi-supervised learning》B Lecouat, C Foo, H Zenati, V Chandrasekhar [A*STAR] (2018) 网页链接 view:网页链接 GitHub:网页链接
《Revisiting the Hierarchical Multiscale LSTM》Á Kádár, M Côté, G Chrupała, A Alishahi [Tilburg University & Microsoft Research Montreal] (2018) 网页链接 view:网页链接
《Learning in Variational Autoencoders with Kullback-Leibler and Renyi Integral Bounds》S Sârbu, R Volpi, A Peşte, L Malagò [Romanian Institute of Science and Technology] (2018) 网页链接 view:网页链接
《BayesGrad: Explaining Predictions of Graph Convolutional Networks》H Akita, K Nakago, T Komatsu, Y Sugawara, S Maeda, Y Baba, H Kashima [Kyoto University & Preferred Networks & University of Tsukuba] (2018) 网页链接 view:网页链接 GitHub:网页链接
《Assessing the Scalability of Biologically-Motivated Deep Learning Algorithms and Architectures》S Bartunov, A Santoro, B A. Richards, G E. Hinton, T Lillicrap [DeepMind & University of Toronto] (2018) 网页链接 view:网页链接
【Learn how to design large-scale systems】网页链接 学习如何设计大规模系统--系统设计入门，为系统设计的面试做准备。被翻译成多种文字，中文翻译：网页链接 学习架构好资源，墙裂建议收藏！
《Scalable Gaussian Processes with Grid-Structured Eigenfunctions (GP-GRIEF)》T W. Evans, P B. Nair [University of Toronto] (2018) 网页链接 view:网页链接 GitHub:网页链接
《A Boo(n) for Evaluating Architecture Performance》O Bajgar, R Kadlec, J Kleindienst [IBM Watson] (2018) 网页链接 view:网页链接
《A network-based citation indicator of scientific performance》C Schulz, B Uzzi, D Helbing, O Woolley-Meza [ETH Zurich & Northwestern University] (2018) 网页链接 view:网页链接
《Sem-GAN: Semantically-Consistent Image-to-Image Translation》A Cherian, A Sullivan [Mitsubishi Electric Research Labs (MERL)] (2018) 网页链接 view:网页链接
《Multi-task Mid-level Feature Alignment Network for Unsupervised Cross-Dataset Person Re-Identification》S Lin, H Li, C Li, A C Kot [University of Warwick & Nanyang Technological University & Charles Sturt University] (2018) 网页链接 view:网页链接
【ACL 2018 Facebook成果汇总】《Facebook Research at ACL 2018》 网页链接
“(colab)Build a linear model with Estimators” 网页链接
“(CMU)NeuLab Presentations at ACL 2018” 网页链接
今日焦点：卷积网络的问题及其解决方案CoordConv——CoordConv解决了坐标变换问题，具有更好的泛化能力，训练速度提高150倍，参数比卷积少10-100倍
《An Intriguing Failing of Convolutional Neural Networks and the CoordConv Solution》R Liu, J Lehman, P Molino... [Uber AI Labs] (2018) 网页链接 paper:网页链接 pytorch implementation of CoordConv GitHub:网页链接
【Python 3教学课程(Jupyter notebooks)】’Jupyter notebooks for teaching/learning Python 3' by Jerry Pussinen GitHub: 网页链接
【IPN:Randal Olson的机器学习初学者实例教程】《An example machine learning notebook》by Randal S. Olson 网页链接 mytinder:网页链接
《Multi-task Mid-level Feature Alignment Network for Unsupervised Cross-Dataset Person Re-Identification》S Lin, H Li, C Li, A C Kot [University of Warwick & Nanyang Technological University & Charles Sturt University] (2018) 网页链接 view:网页链接
【一般感知流形的分类几何】《Classification and Geometry of General Perceptual Manifolds》SY Chung, DD Lee, H Sompolinsky [Harvard University] (2018) 网页链接
今日焦点：生物学启发深度学习算法与架构的可扩展性评估
《Assessing the Scalability of Biologically-Motivated Deep Learning Algorithms and Architectures》S Bartunov, A Santoro, B A. Richards, G E. Hinton, T Lillicrap [DeepMind & University of Toronto] (2018) 网页链接 view:网页链接
【可复现机器学习：软件挑战、八卦和工程解决方案】《Reproducible ML: software challenges, anecdotes and some engineering solutions》by Alexandre Gramfort 网页链接
【哥大课程：面向数据科学的机器学习】《COMS W4721 Machine Learning for Data Science》by John Paisley 网页链接
《Multi-task Mid-level Feature Alignment Network for Unsupervised Cross-Dataset Person Re-Identification》S Lin, H Li, C Li, A C Kot [University of Warwick & Nanyang Technological University & Charles Sturt University] (2018) 网页链接 view:网页链接
《Reflection Analysis for Face Morphing Attack Detection》C Seibold, A Hilsmann, P Eisert [Fraunhofer HHI] (2018) 网页链接 view:网页链接
《A network-based citation indicator of scientific performance》C Schulz, B Uzzi, D Helbing, O Woolley-Meza [ETH Zurich & Northwestern University] (2018) 网页链接 view:网页链接
《Combinatorial Bandits for Incentivizing Agents with Dynamic Preferences》T Fiez, S Sekar, L Zheng, L J. Ratliff [University of Washington] (2018) 网页链接 view:网页链接
《Natural Language Processing for Music Knowledge Discovery》S Oramas, L Espinosa-Anke, F Gómez, X Serra [Universitat Pompeu Fabra & Cardiff University] (2018) 网页链接 view:网页链接
《Using deep learning for comprehensive, personalized forecasting of Alzheimer's Disease progression》C K. Fisher, A M. Smith, J R. Walsh, t C A M Diseases [Unlearn.AI] (2018) 网页链接 view:网页链接
《Manifold regularization with GANs for semi-supervised learning》B Lecouat, C Foo, H Zenati, V Chandrasekhar [A*STAR] (2018) 网页链接 view:网页链接 GitHub:网页链接
《Mask TextSpotter: An End-to-End Trainable Neural Network for Spotting Text with Arbitrary Shapes》P Lyu, M Liao, C Yao, W Wu, X Bai [Huazhong University of Science and Technology & Megvii (Face++) Technology] (2018) 网页链接 view:网页链接
《Revisiting the Hierarchical Multiscale LSTM》Á Kádár, M Côté, G Chrupała, A Alishahi [Tilburg University & Microsoft Research Montreal] (2018) 网页链接 view:网页链接
《Learning in Variational Autoencoders with Kullback-Leibler and Renyi Integral Bounds》S Sârbu, R Volpi, A Peşte, L Malagò [Romanian Institute of Science and Technology] (2018) 网页链接 view:网页链接
《Inducing Grammars with and for Neural Machine Translation》K Tran, Y Bisk [University of Amsterdam & University of Washington] (2018) 网页链接 view:网页链接 GitHub:网页链接
《BayesGrad: Explaining Predictions of Graph Convolutional Networks》H Akita, K Nakago, T Komatsu, Y Sugawara, S Maeda, Y Baba, H Kashima [Kyoto University & Preferred Networks & University of Tsukuba] (2018) 网页链接 view:网页链接 GitHub:网页链接
《Assessing the Scalability of Biologically-Motivated Deep Learning Algorithms and Architectures》S Bartunov, A Santoro, B A. Richards, G E. Hinton, T Lillicrap [DeepMind & University of Toronto] (2018) 网页链接 view:网页链接
今日焦点：GAN技术全景图：损失、架构、正则化和规范化——“Cookbook” for GANs
《The GAN Landscape: Losses, Architectures, Regularization, and Normalization》K Kurach, M Lucic, X Zhai, M Michalski, S Gelly [Google Brain] (2018) 网页链接 view:网页链接 GitHub:网页链接 Compare GAN models Colab 网页链接
【利用Tensorflow对象检测API加速训练与推理】《Accelerated Training and Inference with the Tensorflow Object Detection API | Google AI Blog》 网页链接
在最近举行的SIGIR 2018大会（网页链接）上，机器学习在电子商务领域的应用受到了越来越多的关注，其中值得关注的是来自京东数据科学实验室的Zhaochun Ren（网页链接）、Dawei Yin（网页链接）以及新加坡国立大学的Xiangnan He（网页链接）所带来的讲座“Information Discovery in E-Commerce”（网页链接）。这个讲座涵盖了机器学习在电商领域应用的三个重要方面：搜索、推荐和自然语言处理，着重讲解了一些经典的思路以及最近几年在这些方面的学术动态。
【Tensorflow.js股市预测】《Financial Forecasting using Tensorflow.js (LIVE) - YouTube》by Siraj Raval O网页链接 GitHub:O网页链接 “Tensorflow.js股市预测” 搬运：网页链接
TensorFlow有哪些有意思的接口设计 - 来自知乎专栏「TensorFlow专栏」，作者: tobe TensorFlow有哪些有意思的接口设计
【深度神经网络为什么不易过拟合？傅里叶分析发现固有频谱偏差】过参数化的深度神经网络是一类表达能力极强的函数，甚至能 100% 记住随机数据。但是为什么它们不会轻易地过拟合数据？来自海德堡大学和 MILA 等机构的研究者使用傅立叶分析研究了深度神经网络。网页链接
【教程| 比的Python快100倍，利用spaCy和用Cython实现高速NLP项目】Cython是一个工具包，可以使你在Python中编译C语言，这就是为什么numpy和pandas很快的原因，Cython就是Python的超集。在本文中，作者将为我们介绍他的GitHub项目NeuralCoref v3.0，详解如何利用spaCy和Cython以约100倍于Python的速度实现NLP项目。网页链接