題目：Unsupervised Transfer Learning via Adversarial Contrastive Training

内容簡介：Learning a data representation for downstream supervised learning tasks under unlabeled scenario is both critical and challenging. In this paper, we propose a novel unsupervised transfer learning approach using adversarial contrastive training (ACT). Our experimental results demonstrate outstanding classification accuracy with both fine-tuned linear probe and K-NN protocol across various datasets, showing competitiveness with existing state-of-the-art self-supervised learning methods. Moreover, we provide an end-to-end theoretical guarantee for downstream classification tasks in a misspecified, over-parameterized setting, highlighting how a large amount of unlabeled data contributes to prediction accuracy. Our theoretical findings suggest that the testing error of downstream tasks depends solely on the efficiency of data augmentation used in ACT when the unlabeled sample size is sufficiently large. This offers a theoretical understanding of learning downstream tasks with a small sample size.

報告人：焦雨領

報告人簡介：武漢大學人工智能學院，教授博導，副院長。入選國家高層次青年人才，主要研究機器學習、科學計算。近期關注深度學習數理基礎，在計算數學、應用數學、統計學、電子工程、人工智能等領域的旗艦期刊和會議上發表論文三十多篇：SIAM 系列（5 篇）、Appl.Comput. Harmon. Anal.(2篇)、Inverse Probl. (2 篇)；Ann. Stat. （3 篇）、J.Amer. Statist. Assoc.； IEEE Trans. Inf. Theory (3 篇)、IEEE Trans. Signal Process.（3篇）；J. Mach. Learn. Res. (6 篇)、ICML (3 篇)、NeurIPS (3篇，其中一篇Oral、一篇Spotlight)；Nat. Commun.。主持國家重點研發計劃子課題、國家自然科學基金面上項目及一批同華為開展的校企合作項目。

時  間：2025年4月4日（周五）上午09:30開始

地  點：騰訊會議： 486-526-378

熱烈歡迎廣大師生參加!

 太阳集团1088vip

2025年4月2日