Adversarial Loss

Adversarial loss is a training technique that enhances model robustness by incorporating an adversarial component into the learning process. Current research focuses on applying adversarial loss to diverse machine learning problems, including improving the accuracy and robustness of deep neural networks, mitigating overfitting in generative models, and ensuring fairness in representation learning. This approach is significant because it addresses vulnerabilities in machine learning models to adversarial attacks and improves generalization performance across various domains, leading to more reliable and trustworthy AI systems. The development of efficient algorithms for optimizing adversarial loss, particularly in complex settings like reinforcement learning and continual learning, remains a key area of investigation.

Papers

June 16, 2022

A Closer Look at Smoothness in Domain Adversarial Training
Harsh Rangwani, Sumukh K Aithal, Mayank Mishra, Arihant Jain, R. Venkatesh Babu
Adversarial Training Adversarial Text Glance Annotation Adversarial Loss Domain Adversarial Adversarial Method Average Smoothness

June 9, 2022

Learning to generate imaginary tasks for improving generalization in meta-learning
Yichen Wu, Long-Kai Huang, Ying Wei
LeArning Abstract Strong Generalization Adversarial Loss Meta Training Real World Task Adversarial Data Collection

May 15, 2022

GLaMa: Joint Spatial and Frequency Loss for General Image Inpainting
Zeyu Lu, Junjun Jiang, Junqin Huang, Gang Wu, Xianming Liu
Spatial Domain Adversarial Loss Reconstruction Loss Image Inpainting Frequency Loss

May 7, 2022

Learning Disentangled Textual Representations via Statistical Measures of Similarity
Pierre Colombo, Guillaume Staerman, Nathan Noiry, Pablo Piantanida
LeArning Abstract Deep Neural Network High Similarity Text Encoder Adversarial Loss Textual Representation

April 6, 2022

Masking Adversarial Damage: Finding Adversarial Saliency for Robust and Sparse Network
Byung-Kwan Lee, Junho Kim, Yong Man Ro
Adversarial Robustness Robust Version Adversarial Loss Sparse Network Adversarial Pruning

March 28, 2022

Robust Unlearnable Examples: Protecting Data Against Adversarial Learning
Shaopeng Fu, Fengxiang He, Yang Liu, Li Shen, Dacheng Tao
Adversarial Training Raw Data Adversarial Learning Adversarial Loss Unlearnable Example Error Minimizing Noise

March 15, 2022

Task-Agnostic Robust Representation Learning
A. Tuan Nguyen, Ser Nam Lim, Philip Torr
Adversarial Attack Adversarial Example Robust Representation Adversarial Loss Task Agnostic Representation

February 15, 2022

Between Stochastic and Adversarial Online Convex Optimization: Improved Regret Bounds via Smoothness
Sarah Sachs, Hédi Hadiji, Tim van Erven, Cristóbal Guzmán
Regret Bound Stochastic Way Online Convex Optimization Adversarial Loss Minimax Regret

February 6, 2022

Block shuffling learning for Deepfake Detection
Sitong Liu, Zhichao Lian, Siqi Gu, Liang Xiao
Convolutional Neural Network Deepfake Detection Adversarial Loss Face Forgery Detection Multi Modal Outer Arithmetic Block Deepfake Detection Method

January 25, 2022

Virtual Adversarial Training for Semi-supervised Breast Mass Classification
Xuxin Chen, Ximin Wang, Ke Zhang, Kar-Ming Fung, Theresa C. Thai, Kathleen Moore, Robert S. Mannel, Hong Liu, Bin Zheng, Yuchen Qiu
Adversarial Loss Mammogram Classification Virtual Adversarial Training

January 7, 2022

AugmentedPCA: A Python Package of Supervised and Adversarial Linear Factor Models
William E. Carson, Austin Talbot, David Carlson
Latent Representation Python Library Adversarial Loss Deep Autoencoders

December 17, 2021

Robust Upper Bounds for Adversarial Training
Dimitris Bertsimas, Xavier Boix, Kimberly Villalobos Carballo, Dick den Hertog
Adversarial Attack Adversarial Training Adversarial Loss Robustness Bound

December 9, 2021

HBReID: Harder Batch for Re-identification
Wen Li, Furong Xu, Jianan Zhao, Ruobing Zheng, Cheng Zou, Meng Wang, Yuan Cheng
Loss Function Adversarial Loss Whistleblower Re Identification Triplet Loss Adversarial Generation Example Mining

November 5, 2021

Normalizing Flow as a Flexible Fidelity Objective for Photo-Realistic Super-resolution
Andreas Lugmayr, Martin Danelljan, Fisher Yu, Luc Van Gool, Radu Timofte
Super Resolution Ground Truth Normalizing Flow Adversarial Loss Fidelity Metric Pixel Wise Loss