Simon Guist

Simon Guist is a Doctoral Researcher at the Empirical Inference Department of the Max Planck Institute for Intelligent Systems in Tübingen, Germany. Supervised by Dieter Büchler and Bernhard Schölkopf, his research focuses on improving reinforcement learning efficiency for soft robotics through sim-to-real approaches, particularly for muscle-based robot arms. Prior to his PhD, he completed his Master's thesis at the Max Planck Institute in collaboration with Prof. Jan Peters after studying in Heidelberg.

• Research Interests: Machine Learning, Reinforcement Learning, Soft Robotics, Sim-to-Real Transfer, Neural Networks.
• Contributions: Primary developer of the Hindsight States algorithm and related open-source packages, including stable-baselines3-hindsight-states and Gymnasium-Robotics-HYSR.
• Email: simon.guist@tuebingen.mpg.de
• Profiles: GitHub, LinkedIn, Twitter

His work has been published at Robotics: Science and Systems (RSS) 2023, introducing the Hindsight States algorithm for efficient reinforcement learning through hybrid state simulation. This research aligns with trends in autonomous systems and robotics, leveraging Python and PyTorch frameworks.

Technical Expertise: Python, PyTorch, Docker, Mujoco, Gymnasium, Reinforcement Learning, Soft Robotics.