Yan DAI (戴言)
Hi! I am a second-year PhD student at Operations Research Center (ORC), MIT where I am fortunate to be co-advised by Patrick Jaillet and Negin Golrezaei. In Summer 2026, I am a research intern at Microsoft Research EconCS, mentored by Alex Silvkins and Brendan Lucier; I also work closely with Nicole Immorlica.
Prior to MIT, I earned my Bachelor’s degree from Yao Class, Tsinghua in 2024. I am extremely grateful to Longbo Huang, Haipeng Luo, Simon S. Du, and Suvrit Sra who introduced me to the world of learning theory.
My research focuses on the intersection of Economics and Computer Science (EconCS), specifically tackling strategic behaviors in economic systems via online learning tools. I’m also broadly interested in learning theory: bandits, online learning, reinforcement learning theory, and deep learning theory.
Recent News
- Feb, 2026. Named as a Runner-Up of Two Sigma PhD Fellowship (2 winners and 2 runners-up in 2026).
- Sep, 2025. Incentive-Aware Dynamic Resource Allocation under Constraints accepted to NeurIPS 2025.
- Jun, 2025. Adversarial Network Optimization under Bandit Feedback won Best Paper Award at ACM SIGMETRICS 2025!
- Jun, 2025. Incentive-Aware Dynamic Resource Allocation under Constraints won 1st Place in ACM Student Research Competition (SRC) held at SIGMETRICS 2025!
- May, 2025. Non-Monetary Mechanism Design without Distributions accepted to COLT 2025.
- Jan, 2025. Refined Best-of-Both-Worlds Heavy-Tailed MABs accepted to ICLR 2025 as Spotlight (5.1%).
- Dec, 2024. Adversarial Network Optimization under Bandit Feedback published in POMACS journal.
- Jun, 2024. Awarded Outstanding Graduate by Beijing, by Tsinghua, and by Yao Class.
- May, 2024. Refined Linear Markov Games accepted to COLT 2024.
- May, 2024. Adam is FTRL in Disguise accepted to ICML 2024.
- Nov, 2023. Awarded Presidential Scholarship (10 undergrads per year) by Tsinghua!
- Sep, 2023. Role of Normalization in SAM accepted to NeurIPS 2023.
- Sep, 2023. Awarded “Andrew C. Yao Award” Gold Medal (1 per year) by Yao Class!
- Apr, 2023. Refined Adversarial Linear(-Q) MDPs and Banker-OMD Framework for Delayed Bandits accepted to ICML 2023.
- Jan, 2023. Variance-Aware Sparse Linear Bandits accepted to ICLR 2023.
- Sep, 2022. FTPL in Adversarial MDPs accepted to NeurIPS 2022.
- May, 2022. Best-of-Both-Worlds Heavy-Tailed MABs accepted to ICML 2022.
Professional Services
- Journal Reviewing: Journal of Machine Learning Research (JMLR), IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), Machine Learning; IEEE/ACM Transactions on Networking (TON), Performance Evaluation (PEVA), IEEE Internet of Things Journal (IoT-J).
- Conference Reviewing: COLT 2025, ICML 2026 (Gold reviewer) / 2025 / 2023, NeurIPS 2026 / 2025 / 2024 / 2023 / 2022, ICLR 2025 / 2024, AISTATS 2026 / 2025 / 2023 / 2022, AAAI 2026, ALT 2023.