Yongyi Mao

Full Professor@University of Ottawa·加拿大·OpenReview

研究方向

Machine Learning · Information Theory · Communications · Learning theory; deep learning theory · techniques and applications · Coding and information theory

4.5

Two Facets of SDE Under an Information-Theoretic Lens: Generalization of SGD via Training Trajectories and via Terminal States

ICLR 2024Rejected

二作

4.8

On robust overfitting: adversarial training induced distribution matters

合作者 (5)

Yongyi Mao

Universally Optimal Watermarking Schemes for LLMs: from Theory to Practice

On $f$-Divergence Principled Domain Adaptation: An Improved Framework

Generalization Bounds via Conditional $f$-Information

Two Facets of SDE Under an Information-Theoretic Lens: Generalization of SGD via Training Trajectories and via Terminal States

On robust overfitting: adversarial training induced distribution matters