Adversarial Discriminator

Adversarial discriminators are components of machine learning models designed to distinguish between real and artificially generated data, playing a crucial role in training generative models and enhancing robustness against adversarial attacks. Current research focuses on improving discriminator architectures and training methods, including the use of diffusion models, dynamic label strategies, and multi-layer designs, to address issues like unstable training, mode collapse, and suboptimal performance. These advancements are significant for improving the performance and reliability of various applications, ranging from object detection and natural language processing to image compression and solving partial differential equations. The ongoing development of more effective and efficient adversarial discriminators is driving progress across numerous machine learning subfields.

Papers

September 11, 2024

AdvLogo: Adversarial Patch Attack against Object Detectors based on Diffusion Models
Boming Miao, Chunxiao Li, Yao Zhu, Weixiang Sun, Zizhe Wang, Xiaoyi Wang, Chuanlong Xie
Diffusion Model Object Detector Adversarial Patch Adversarial Patch Attack Adversarial Region Adversarial Discriminator

August 23, 2024

Dynamic Label Adversarial Training for Deep Learning Robustness Against Adversarial Attacks
Zhenyu Liu, Haoran Duan, Huizhi Liang, Yang Long, Vaclav Snasel, Guiseppe Nicosia, Rajiv Ranjan, Varun Ojha
Adversarial Attack Adversarial Example Adversarial Training Model Robustness Neural Network Robustness Adversarial Distillation Adversarial Discriminator

August 15, 2024

KOALA: Enhancing Speculative Decoding for LLM via Multi-Layer Draft Heads with Adversarial Learning
Kaiqi Zhang, Jing Zhao, Rui Chen
Medical LLM Adversarial Learning Speculative Decoding Inference Latency Non Autoregressive Adversarial Discriminator

May 9, 2024

Hard Work Does Not Always Pay Off: Poisoning Attacks on Neural Architecture Search
Zachary Coalson, Huazheng Wang, Qingyun Wu, Sanghyun Hong
Neural Architecture Search Poisoning Attack Manual Effort Clean Label Attack Adversarial Discriminator

February 26, 2024

Training Implicit Generative Models via an Invariant Statistical Loss
José Manuel de Frutos, Pablo M. Olmos, Manuel A. Vázquez, Joaquín Míguez
Generative Adversarial Network Implicit Generative Invariant Loss Adversarial Discriminator

January 26, 2023

Improving Statistical Fidelity for Neural Image Compression with Implicit Local Likelihood Models
Matthew J. Muckley, Alaaeldin El-Nouby, Karen Ullrich, Hervé Jégou, Jakob Verbeek
Image Compression Neural Image Compression Lossy Image Compression Distortion Metric Vq Vae Adversarial Discriminator

November 28, 2022

Lightning Fast Video Anomaly Detection via Adversarial Knowledge Distillation
Nicolae-Catalin Ristea, Florinel-Alin Croitoru, Dana Dascalescu, Radu Tudor Ionescu, Fahad Shahbaz Khan, Mubarak Shah
Anomaly Detection Video Anomaly Anomaly Map Adversarial Distillation Adversarial Discriminator Adversarial Knowledge Distillation

July 14, 2022

PIAT: Physics Informed Adversarial Training for Solving Partial Differential Equations
Simin Shekarpaz, Mohammad Azizmalayeri, Mohammad Hossein Rohban
Adversarial Attack Adversarial Training Partial Differential Equation Theoretical Physic Adversarial Discriminator Physic Based Adversarial

July 13, 2022

Adversarially-Aware Robust Object Detector
Ziyi Dong, Pengxu Wei, Liang Lin
Adversarial Robustness Adversarial Image Robust Detection Robust Object Detection Adversarial Discriminator

June 23, 2022

Disentangling representations in Restricted Boltzmann Machines without adversaries
Jorge Fernandez-de-Cossio-Diaz, Simona Cocco, Remi Monasson
Adversarial Training Meaningful Representation Adversary Agent Restricted Boltzmann Machine Unsupervised Machine Learning Adversarial Discriminator Simple Generative

Adversarial Discriminator

Papers

AdvLogo: Adversarial Patch Attack against Object Detectors based on Diffusion Models

Dynamic Label Adversarial Training for Deep Learning Robustness Against Adversarial Attacks

KOALA: Enhancing Speculative Decoding for LLM via Multi-Layer Draft Heads with Adversarial Learning

Hard Work Does Not Always Pay Off: Poisoning Attacks on Neural Architecture Search

Training Implicit Generative Models via an Invariant Statistical Loss

Improving Statistical Fidelity for Neural Image Compression with Implicit Local Likelihood Models

Lightning Fast Video Anomaly Detection via Adversarial Knowledge Distillation

PIAT: Physics Informed Adversarial Training for Solving Partial Differential Equations

Adversarially-Aware Robust Object Detector

Disentangling representations in Restricted Boltzmann Machines without adversaries