Tianhao Wang (王天浩)
Ph.D. student, Department of Statistics and Data Science, Yale University.
Department of S&DS
219 Prospect Avenue
New Haven, CT 06511
tianhao.wang@yale.edu
I am a final year Ph.D. student in the Department of Statistics and Data Science at Yale University. I am very fortunate to be advised by Prof. Zhou Fan. I am broadly interested in various aspects of statistics and machine learning theory.
Prior to Yale, I obtained my Bachelor’s degree in mathematics with a dual degree in computer science at University of Science and Technology of China.
CVRecent papers(*: equal contribution)
Foundations of Transformers
-
Implicit regularization of gradient flow on one-layer softmax attentionarXiv:2403.08699, 2024Accepted to ICLR 2024 Workshop on Bridging the Gap Between Practice and Theory in Deep Learning
-
How well can Transformers emulate in-context Newton's method?arXiv:2403.03183, 2024Accepted to ICLR 2024 Workshop on Bridging the Gap Between Practice and Theory in Deep Learning
-
Training dynamics of multi-head softmax attention for in-context learning: emergence, convergence, and optimalityarXiv:2402.19442, 2024Accepted to ICLR 2024 Workshop on Bridging the Gap Between Practice and Theory in Deep Learning
Approximate Message Passing algorithms
Implicit bias of optimization algorithms
-
Fast mixing of stochastic gradient descent with normalization and weight decayIn Advances in Neural Information Processing Systems (NeurIPS), 2022
-
Implicit bias of gradient descent on reparametrized models: On equivalence to mirror descentIn Advances in Neural Information Processing Systems (NeurIPS), 2022Abridged version accepted for a contributed talk to ICML 2022 Workshop on Continuous time methods for machine learning
Data-driven decision-making problems
-
Noise-adaptive Thompson sampling for linear contextual banditsIn Advances in Neural Information Processing Systems (NeurIPS), 2023