My research ambition is to build autonomous agents that can solve a wide variety of complex tasks and continuously learn new ones. I believe this calls for decision-making systems that can effectively build on prior knowledge (e.g. large-scale pre-training) and use that prior to continually generalize to new tasks quickly. Towards this goal, I work on various aspects of robotics and reinforcement learning. Currently, I am most excited about incorporating RL into large-scale pre-training for robotic foundation models, and how to post-train robotic policies to extremely high performance with and without RL. I am fortunate to be advised by professor Sergey Levine at UC Berkeley. During my undergrad I was advised by professors George Konidaris and Michael Littman at Brown. Please check out my selected work below.

CV | Google Scholar | Github | Twitter

Publications

Conferences / Workshops

If we have tons of compute to spend to train value functions, how can we be sure we're spending it optimally? In our new paper, we analyze the interplay of model size, UTD, and batch size for training value functions achieving optimal performance.https://t.co/gIF4tzvENS pic.twitter.com/MQzPjpHA5X — Preston Fu (@preston_fu) September 2, 2025

Compute-Optimal Scaling for Value-Based Deep RL RL, scaling laws Preston Fu, Oleh Rybkin, Zhiyuan Zhou, Michal Nauman, Peter Abbeel, Sergey Levine, Aviral Kumar NeurIPS, 2025 [paper] Big models (TD-)overfit less and produce better TD targets, and allows for using bigger batch sizes. We analyze the interplay between model size, batch size, and UTD for scaling value-based RL, and show scaling laws that shows the budget-optimal data-compute tradeoff.

Behavioral Exploration: Learning to Explore via In-Context Adaptation In-context learning Andrew Wagenmaker, Zhiyuan Zhou, Sergey Levine International Conference on Machine Learning (ICML), 2025 [paper] Behavioral exploration (BE) seeks to train policies that can explore over the space of expert demonstration behaviors. We achieve this by training a long-context policy that's conditioned on history and a notion of "coverage-to-go".

Everyone knows action chunking is great for imitation learning. It turns out that we can extend its success to RL to better leverage prior data for improved exploration and online sample efficiency! https://t.co/J5LdRRYbSH The recipe to achieve this is incredibly simple. 🧵 1/N pic.twitter.com/c2N7PZL9Pw — Qiyang Li @ ICML (@qiyang_li) July 12, 2025

Reinforcement Learning with Action Chunking deep reinforcement learning Qiyang Li, Zhiyuan Zhou, Sergey Levine NeurIPS, 2025 [paper] [website] [code] Q-chunking runs RL on a temporally extended action (action chunking) space with an expressive behavior constraint to leverage prior data for improved exploration and online sample efficiency.

AutoEval: Autonomous Evaluation of Generalist Robot Manipulation Policies in the Real World Robotics Zhiyuan Zhou, Pranav Atreya, You Liang Tan, Karl Pertsch, Sergey Levine Conference on Robot Learning (CoRL), 2025. ICLR Robot Learning Workshop, 2025. (Oral) RSS Robot Evaluation Workshop, 2025. (Best Paper) [paper] [website] [code] Evaluating generalist robot policies is very time consuming. We propose AutoEval, a system that autonomously evaluates generalist policies in the real world with learned success detectors and reset policies. We open access to two AutoEval stations for researchers to evaluate their policies. Submit your policy for evaluation in minutes!

Learning Transferable Sub-Goals by Hypothesizing Generalizing Features hierarchical RL skill transfer Anita de Mello Koch, Akhil Bagaria, Bingnan Huo, Zhiyuan Zhou, Cameron Allen, George Konidaris AAAI Workshop on Generalization in Planning, 2025. (Oral) [paper] How do we transfer low level skills in hierarchical RL to make it sample efficient, especially in the context of one-shot and few-shot learning? We find that we can learn subgoals classifiers that transfer to unseen tasks with an ensemble of classifiers that focus on diverse features.

Efficient Online Reinforcement Learning Fine-Tuning Need Not Retain Offline Data deep reinforcement learning fine-tuning Zhiyuan Zhou*, Andy Peng*, Qiyang Li, Sergey Levine, Aviral Kumar International Conference on Learning Representations (ICLR), 2025 [paper] [website] [code] Can we finetune policies and values from offline RL *without retaining the offline data*? Current methods require keeping the offline data for stability and performance, but this make RL hard to scale up when the offline dataset gets bigger and bigger. Turns out a simple recipe, Warm-start RL, is able to finetune rapidly without data retention!

Autonomous Improvement of Instruction Following Skills via Foundation Models robotics autonomous improvement language-conditioned skills VLM Zhiyuan Zhou*, Pranav Atreya*, Abraham Lee, Homer Walke, Oier Mees, Sergey Levine Conference on Robot Learning (CoRL), 2024. [website] [arXiv] [code] [dataset] Can robots self-improve by collecting data autonomously🤖? We introduce SOAR, a system for large-scale autonomous data collection 🚀 and autonomous improvement📈of a multi-task language-conditioned policy in diverse scenes without human interventions .

Tiered Reward: Designing Rewards for Specification and Fast Learning of Desired Behavior behavior specification reward design Pareto optimality fast learning Zhiyuan Zhou, Shreyas Sundara Raman, Henry Sowerby, Michael Littman Reinforcement Learning Conference (RLC), 2024. [paper] [website] [code] [thread] Do you need a reward function for your goal-reaching task? Use Tiered Reward! We prove that Tiered Reward guarantees to lead to an optimal policy, and show that it can lead to fast learning in various deep and tabular environments.

Characterizing the Action-Generalization Gap in Deep Q-Learning action generalization DQN Zhiyuan Zhou, Cameron Allen, Kavosh Asadi, George Konidaris Multidisciplinary Conference on Reinforcement Learning and Decision Making (RLDM), 2022. [arXiv] [poster] [code] We introduce a way to evaluate action-generalization in Deep Q-Learning using an oracle (expert knowledge of action similarity), and shows that DQN's ability to generalize over actions depends on the size of the action space.

Designing Rewards for Fast Learning reward design Interactive RL Henry Sowerby, Zhiyuan Zhou, Michael Littman Multidisciplinary Conference on Reinforcement Learning and Decision Making (RLDM), 2022. (Oral) [arXiv] [poster] [oral at RLDM at 1:20:00] What kind of reward functions make RL fast? We advocate for rewards with big action gaps and small "subjective discounts". We present an algorithm to design these rewards.

School Journal

Policy Transfer in Lifelong Reinforcement Learning through Learning Generalizing Features lifelong RL transfer learning attention Zhiyuan Zhou (Advisor: George Konidaris) Undergraduate Honors Thesis, Brown CS, 2023. [pdf] [code] Introduces an approach to learn state features that generalize across tasks drawn from the same distribution. We use an attantion mechanism to learn an ensemble of minimally overlapping state features, leading to an ensemble of policies. We then use a bandit algorithm to learn to identify the generalizing feature in the ensemble and capitalize on that to learn a transferable policy.

Improving Post-Processing on Video Object Recognition Using Inertial Measurement Unit object recognition Hidden Markov Models Kalman Filter Inertial Measurement Unit Zhiyuan Zhou, Spencer Boyum, Michael Paradiso Brown Undergraduate Research Journal, Spring 2022. [paper on page 29] [code] How to improve the accuracy of object recognition in videos if given per-frame inertial measurements of the camera. We propose two way to do so.