Ruoqi Zhang

Stay Huuuungry Stay Foooolish

prof_pic.jpg

ÅNG 103175 hus 10

Lägerhyddsvägen 1

Box 337 751 05 UPPSALA

I am a PhD Candidate supervised by Per Mattsson and co-supervised by Torbjörn Wigren at Uppsala University. Beofore my halftime, my research focuses on Reinforcement Learning (RL) and Automatic Control, particularly addressing how control could assist RL and uncertainty in RL. Now I’m more into fiffusion-based generative models in both offeline and offline-to-online RL.

I earned my master from Uppsala Univerity in 2020 and did my thesis in Uppsala System and Control Division. Beyond my research, I organize SysCon RL Reading Group, engaging weekly in recent RL papers to foster understanding and innovation within the field. Explore past discussions here.

news

Mar 15, 2025 🎉 We have our first milestone collabrating with EA! We utilize the consistency model as the policy for fast inference on game engine and improve the performance via q-ensembels. Check the paper via Real-Time Diffusion Policies for Games: Enhancing Consistency Policies with Q-Ensembles.
Oct 21, 2024 🎉 I will visit Professor Dominik Bauman who is leading Cyber-physical Systems Group from Oct 22nd to Dec 5th at Aalto University. This visit is partially funded by Liljewalch travel scholarships.
Sep 25, 2024 🎉 Our work Entropy-regularized Diffusion Policy with Q-Ensembles for Offline Reinforcement Learning is accepted by Neurips 2024, Vancouver.
Jul 24, 2024 🎉 Our paper Safe Output Feedback Improvement with Baselines was accpted by IEEE CDC 2024 Milano.
May 14, 2024 🎉 I will attend ETPL-ETHZ Multi-Agent Reinforcement Learning from July 29th to July 31st in Lausanne.

selected publications

  1. Real-Time Diffusion Policies for Games: Enhancing Consistency Policies with Q-Ensembles
    Zhang, Ruoqi, Luo, Ziwei, Sjölund, Jens, Mattsson, Per, Gisslén, Linus, and Sestini, Alessandro
    arXiv preprint arXiv:2503.16978 2025
  2. Entropy-regularized Diffusion Policy with Q-Ensembles for Offline Reinforcement Learning
    Zhang, Ruoqi, Luo, Ziwei, Sjölund, Jens, Schön, Thomas B., and Mattsson, Per
    In Advances in Neural Information Processing Systems 2024
  3. Observer-Feedback-Feedforward Controller Structures in Reinforcement Learning
    Zhang, Ruoqi, Mattsson, Per, and Wigren, Torbjörn
    IFAC-PapersOnLine 2023 22nd IFAC World Congress, Yokohama, Japan
  4. Robust nonlinear set-point control with reinforcement learning
    Zhang, Ruoqi, Mattsson, Per, and Wigren, Torbjörn
    In American Control Conference, ACC 2023, San Diego, CA, USA, May 31 - June 2, 2023 2023
  5. Risk-sensitive Actor-free Policy via Convex Optimisation
    Zhang, Ruoqi, and Sjölund, Jens
    In Proceedings of the IJCAI-23 Joint Workshop on Artificial Intelligence Safety and Safe Reinforcement Learning , Macau, China, August 21-22, 2023 2023