学习笔记：【VALSE短教程】《Adversarial Attack and Defense》
学习笔记：【VALSE短教程】《Adversarial Attack and Defense》
视频地址
1、White-box attacks
Direction I
论文地址：
EXPLAINING AND HARNESSING ADVERSARIAL EXAMPLES
论文地址：
ADVERSARIAL EXAMPLES IN THE PHYSICAL WORLD
论文地址：
Towards Deep Learning Models Resistant to Adversarial Attacks
Direction II
分错条件下找到扰动最小的对抗样本。
论文地址：
DeepFool: a simple and accurate method to fool deep neural networks
论文地址：
Towards Evaluating the Robustness of Neural Networks
2、Black-box Attack
1、Transferability-based Attack
论文地址：
Boosting Adversarial Attacks with Momentum
NESTEROV ACCELERATED GRADIENT AND SCALE INVARIANCE FOR ADVERSARIAL ATTACKS
论文地址：
Towards Understanding and Improving the Transferability of Adversarial Examples in Deep Neural Networks
Evading Defenses to Transferable Adversarial Examples by Translation-Invariant Attacks
Improving Transferability of Adversarial Examples with Input Diversity
论文地址：
SKIP CONNECTIONS MATTER: ON THE TRANSFERABILITY OF ADVERSARIAL EXAMPLES GENERATED WITH RESNETS
2、Query-based Adversarial Attack
论文地址：
DECISION-BASED ADVERSARIAL ATTACKS: RELIABLE ATTACKS AGAINST BLACK-BOX MACHINE LEARNING MODELS
A Ray Searching Method for Hard-label Adversarial Attack
ZOO: Zeroth Order Optimization Based Black-box Attacks to Deep Neural Networks without Training Substitute Models
Auto ZOOM: Autoencoder-based Zeroth Order Optimization Method for Attacking Black-box Neural Networks
Square Attack: a query-efficient black-box adversarial attack via random search
Black-box Adversarial Attacks with Limited Queries and Information
N ATTACK:Learning the Distributions of Adversarial Examples for an Improved Black-Box Attack on Deep Neural Networks
Improving Query Efficiency of Black-box Adversarial Attack
论文地址：
Adversarial Patch
Robust Physical-World Attacks on Deep Learning Visual Classification
论文地址：
Adversarial Camouflage: Hiding Physical-World Attacks with Natural Styles
论文地址：
BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain
论文地址：
Towards Deep Learning Models Resistant to Adversarial Attacks
论文地址：
On the Convergence and Robustness of Adversarial Training
论文地址：
Adversarial Weight Perturbation HelpsRobust Generalization
论文地址：
Visualizing the Loss Landscape of Neural Nets
论文地址：
Understanding Adversarial Robustness Through Loss Landscape Geometries
INTERPRETING ADVERSARIAL ROBUSTNESS:A VIEW FROM DECISION SURFACE IN INPUT SPACE
论文地址：
Theoretically Principled Trade-off between Robustness and Accuracy
论文地址：
IMPROVING ADVERSARIAL ROBUSTNESS REQUIRES REVISITING MISCLASSIFIED EXAMPLES
论文地址：
Adversarial Neuron Pruning Purifies Backdoored Deep Models
论文地址：
Spectral Signatures in Backdoor Attacks
DEEP PARTITION AGGREGATION:PROVABLE DEFENSES AGAINST GENERAL POISONING
ATTACKS
Data Poisoning against Differentially-Private Learners: Attacks and Defenses
STRONG DATA AUGMENTATION SANITIZES POISONING AND BACKDOOR ATTACKS WITHOUT AN ACCURACY TRADEOFF
Neural Cleanse: Identifying and Mitigating Backdoor Attacks in Neural Networks
Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networks
REFIT: A Unified Watermark Removal Framework For Deep Learning Systems With Limited Data
论文地址：
Feature Denoising for Improving Adversarial Robustness
论文地址：
Adversarial Examples Improve Image Recognition
论文地址：
Improving Adversarial Robustness via Channel-wise Activation Suppressing
论文地址：
Implicit Euler Skip Connections: Enhancing Adversarial Robustness via Numerical Stability
论文地址：
Unlearnable Examples: Making Personal Data Unexploitable
论文地址：
Unadversarial Examples: Designing Objects for Robust Vision
