Adversarial Transfer

Adversarial transfer learning focuses on improving the robustness and generalizability of machine learning models, particularly in scenarios with limited data or significant domain discrepancies. Current research emphasizes techniques like adversarial training, domain assimilation, and carefully designed model initialization strategies (e.g., robust linear probing) to mitigate the negative effects of adversarial attacks and domain shifts, often within the context of transfer learning from large pre-trained models. This work is crucial for enhancing the reliability and trustworthiness of AI models across diverse applications, including medical image analysis and reinforcement learning, where robustness is paramount. The ultimate goal is to create more reliable and generalizable AI systems that are less susceptible to manipulation and perform well even with limited training data.

Papers

August 26, 2024

Histology Virtual Staining with Mask-Guided Adversarial Transfer Learning for Tertiary Lymphoid Structure Detection
Qiuli Wang, Yongxu Liu, Li Ma, Xianqi Wang, Wei Chen, Xiaohong Yao
Immunohistochemistry Image Virtual Staining Virtual Immunohistochemistry Lymphocyte Detection Adversarial Transfer

February 25, 2024

Adversarial-Robust Transfer Learning for Medical Imaging via Domain Assimilation
Xiaohui Chen, Tie Luo
Artificial Intelligence Adversarial Attack Transfer Learning Medical Image Medical Imaging Trustworthy Deep Learning Adversarial Transfer Selective Knowledge Assimilation

December 10, 2023

Initialization Matters for Adversarial Transfer Learning
Andong Hua, Jindong Gu, Zhiyu Xue, Nicholas Carlini, Eric Wong, Yao Qin
Adversarial Robustness New Initialization Finetune Paradigm Adversarial Fine Tuning Robust Pre Adversarial Transfer

August 29, 2023

Adversarial Style Transfer for Robust Policy Optimization in Deep Reinforcement Learning
Md Masudur Rahman, Yexiang Xue
Reinforcement Learning Deep Reinforcement Learning Reinforcement Learning Agent Robust Policy Adversarial Reinforcement Learning Deep Reinforcement Learning Algorithm Adversarial Transfer

May 30, 2023

October 6, 2022

Adversarial Lagrangian Integrated Contrastive Embedding for Limited Size Datasets
Amin Jalali, Minho Lee
Augmentation Method Contrastive Training Adversarial Contrastive Learning Small Scale Datasets Adversarial Transfer

September 1, 2022

Adversarial Stain Transfer to Study the Effect of Color Variation on Cell Instance Segmentation
Huaqian Wu, Nicolas Souedet, Camille Mabillon, Caroline Jan, Cédric Clouchoux, Thierry Delzescaux
Mixed Effect Instance Segmentation Cell Segmentation Stain Normalization Stain Variation Adversarial Transfer Color Variation

May 25, 2022

Low Resource Style Transfer via Domain Adaptive Meta Learning
Xiangyang Li, Xiang Long, Yu Xia, Sujian Li
Adversarial Training Unsupervised Text Style Transfer Adversarial Transfer

Adversarial Transfer

Papers

Histology Virtual Staining with Mask-Guided Adversarial Transfer Learning for Tertiary Lymphoid Structure Detection

Adversarial-Robust Transfer Learning for Medical Imaging via Domain Assimilation

Initialization Matters for Adversarial Transfer Learning

Adversarial Style Transfer for Robust Policy Optimization in Deep Reinforcement Learning

Multi-source adversarial transfer learning for ultrasound image segmentation with limited similarity

Multi-source adversarial transfer learning based on similar source domains with local features

Adversarial Lagrangian Integrated Contrastive Embedding for Limited Size Datasets

Adversarial Stain Transfer to Study the Effect of Color Variation on Cell Instance Segmentation

Low Resource Style Transfer via Domain Adaptive Meta Learning