ICML 2025
About
Hi there! I am a machine learning researcher interested in both the foundations of learning and their real-world applications. My research interests include optimization for machine learning, reinforcement learning, and mathematical optimization theory.
I am currently an Applied Scientist at Amazon on the Prime Video Personalization team, where I work on whole-page reranking. My earlier industry experience includes conversational recommendation at Amazon Prime Video and ads ranking models at TikTok Mall.
Before joining Amazon, I pursued my Ph.D. in the Department of Computer Science at Purdue University, advised by Prof. Brian Bullins. My doctoral research gave me a strong foundation in mathematical optimization theory and machine learning. I have also worked with Prof. Raymond A. Yeh on AI safety, including model immunization and machine unlearning, and Prof. Jean Honorio on statistical learning theory.
I received an M.S. in Statistics and Computer Science from Purdue in 2024 alongside my Ph.D. studies, and a B.E. in Computer Science from Xi’an Jiaotong University in 2020, where my research focused on deep reinforcement learning and robotics.
Affiliations
Recent News
- Our work on multi-turn LLM conversational recommendation was accepted to CIKM 2026.
- Joined Amazon as an Applied Scientist in Sunnyvale, California.
- Graduated with a Ph.D. in Computer Science from Purdue University.
- Our paper on variance adaptive optimizer for LLM pretraining was accepted to ICML 2026.
- Defended my thesis, Algorithms and Theory for Optimization with Structured Objectives.
- Reviewer for JMLR.
- Reviewer for NeurIPS 2026.
- Reviewer for ICML 2026.
- Reviewer for ICLR 2026.
- Reviewer for TPAMI.
- Our paper on model immunization through condition number was accepted to ICML 2025 (Oral, top 1%). See you in Vancouver!
- Started summer internship as an applied scientist at Amazon Prime Video, Seattle, Washington.
- Reviewer for NeurIPS 2025.
- Our paper on acceleration for ℓp steepest descent was accepted to COLT 2025.
- Our paper on stochastic ℓp descent for nonconvex optimization was accepted to ICML 2025.
- Our paper on high-order & uniformly convex optimization was accepted to ICLR 2025 (Oral, top 1.8%).
- Reviewer for ICML 2025.
- Passed the Ph.D. preliminary exam, counting down now.
- Reviewer for AISTATS 2025.
- Reviewer for TMLR.
- Reviewer for ICLR 2025.
- Reviewer for AAAI 2025.
- Summer internship as a machine learning engineer at TikTok, San Jose, California.
- Reviewer for NeurIPS 2024.
- Received M.S. in Statistics and Computer Science along the way of my doctoral journey!
- Our paper on federated composite saddle point optimization was accepted to ICLR 2024.
- Our paper on dual convexified CNNs was accepted to Transactions on Machine Learning Research (TMLR) 2024.
- Reviewer for ICML 2024.
- Reviewer for ICLR 2024.
- Reviewer for NeurIPS 2023.
- Reviewer for AISTATS 2023.
- Passed Algorithm and System core courses (equivalence of qualification exam in Purdue CS).
- Our paper on hindsight deep reinforcement learning was accepted to IJCAI 2021.
- Admitted to fully funded CS Doctoral Program from Purdue University.
- A marvelous week with Prof. David Hsu from NUS during his visit to our group.
- Wonderful time at Venetian Hotel, Macau, China attending IROS 2019.
- Two papers I coauthored were accepted to IROS 2019.
Scroll for earlier news.