Augmented Domain Adaptation

Augmented domain adaptation tackles the challenge of transferring knowledge learned from one dataset (source domain) to a different, but related, dataset (target domain) where labeled data is scarce or absent. Current research focuses on improving this transfer by augmenting the source domain with synthetic data or by leveraging techniques like in-context learning and meta-learning to bridge the distributional gap between domains. These methods, often implemented using neural networks and incorporating adversarial learning or proxy-based approaches, aim to enhance the robustness and accuracy of models across diverse data distributions. This work has significant implications for various applications, including improving the generalizability of machine learning models in computer vision, natural language processing, and biomedical signal processing.

Papers

January 1, 2024

Towards Improved Proxy-based Deep Metric Learning via Data-Augmented Domain Adaptation
Li Ren, Chen Chen, Liqiang Wang, Kien Hua
Domain Adaptation Deep Metric Learning Distance Metric Proxy Metric Proxy Based Deep Metric Learning Augmented Domain Adaptation

November 20, 2023

Adapt in Contexts: Retrieval-Augmented Domain Adaptation via In-Context Learning
Quanyu Long, Wenya Wang, Sinno Jialin Pan
Large Language Model Domain Adaptation Context Learning Context Information Augmented Domain Adaptation

October 20, 2023

Towards Subject Agnostic Affective Emotion Recognition
Amit Kumar Jaiswal, Haiming Liu, Prayag Tiwari
Domain Adaptation Emotion Recognition Brain Computer Interface Augmented Domain Adaptation

December 27, 2022

Robust Cross-vendor Mammographic Texture Models Using Augmentation-based Domain Adaptation for Long-term Breast Cancer Risk
Andreas D. Lauritzen, My Catarina von Euler-Chelpin, Elsebeth Lynge, Ilse Vejborg, Mads Nielsen, Nico Karssemeijer, Martin Lillholm
Cancer Screening Breast Cancer Risk Augmented Domain Adaptation

February 21, 2022

Domain-Augmented Domain Adaptation
Qiuhao Zeng, Tianze Luo, Boyu Wang
Domain Adaptation Unsupervised Domain Adaptation Unlabeled Target Domain Domain Discrepancy Augmented Domain Adaptation

Augmented Domain Adaptation

Papers

Towards Improved Proxy-based Deep Metric Learning via Data-Augmented Domain Adaptation

Adapt in Contexts: Retrieval-Augmented Domain Adaptation via In-Context Learning

Towards Subject Agnostic Affective Emotion Recognition

Robust Cross-vendor Mammographic Texture Models Using Augmentation-based Domain Adaptation for Long-term Breast Cancer Risk

Domain-Augmented Domain Adaptation