Having dreams is what makes life tolerable.
Hi there! I am Zilin, a Computer Science DPhil student at the University of Oxford in the WhiRL and FLAIR labs. I am fortunate to be co-supervised by Professor Shimon Whiteson and Professor Jakob Foerster. I previously received a Master’s degree from Tsinghua University, supervised by Prof. Zhiyong Wu. Before that, I did my undergraduate studies at Dalian University of Technology.
My research focuses on efficient and real-world impactful Deep Reinforcement Learning (RL), with an emphasis on how to improve pretrained control policies with RL post-training.
$$
\small\texttt{Email: [email protected]}
$$
**CV / Google Scholar / Github / X / Linkedin**
News
- April 30, 2026 - Our paper Procedural Generation of Algorithm Discovery Tasks in Machine Learning is accepted by ICML 2026.
- April 01, 2026 - Visited Waymo and gave a talk about NOMAD.
- Mar 10, 2026 - Attended Isambard Summit and gave a talk about NOMAD.
- Mar 10, 2026 - Gave a talk at Motional about NOMAD.
- Feb 06, 2026 - Gave a talk at EMERGE lab about NOMAD.
- Nov 27, 2025 - We release the code of DiscoBench: An Open-Ended Benchmark For Algorithm Discovery, check it!
- Aug 07, 2025 - Our paper How Should We Meta-Learn Reinforcement Learning Algorithms? was awarded Outstanding Paper for Scientific Understanding In Reinforcement Learning at RLC 2025. Big congrats to all the authors!
- May 09, 2025 - Our work How Should We Meta-Learn Reinforcement Learning Algorithms? is accepted by RLC 2025.
- May 01, 2025 - Our paper Network Sparsity Unlocks the Scaling Potential of Deep Reinforcement Learning is accepted for an oral presentation (Top ~1%) at ICML 2025.
- Oct 05, 2024 - Started my DPhil at Oxford!
Selected Research
Personal Interests