GitHub - ccli3896/RLWorms: Setting up C elegans as an RL environment.

This repository contains code and data used to set up the nematode C elegans as an RL environment. To accompany paper Discovering neural policies to drive behavior by integrating deep reinforcement learning agents with biological neural networks by Li, Kreiman, and Ramanathan (2023). Guiding or improving animal behavior directly through the nervous system has been a common goal for neuroscience and robotics researchers alike [1-3]. Previous works in brain interfaces and animal robotics have attempted to use direct interventions to affect behavior on a variety of tasks, relying on manual specification for stimulation frequencies, locations, dynamics, and patterns [4–21]. A central difficulty with these approaches is that manual tuning has limited applicability, as it relies on knowledge of the neural circuits or mechanisms involved. For direct neural stimulation, effective patterns can vary depending on which neurons are targeted and on the animal itself [22,23], and thus, even though t

saved by

• Yuxiang Wang

related reading

• "the digital sphinx: can a worm brain control a fly body?" TheSphinx_2026.pdffaculty.washington.edu
• Debugging Reinforcement Learning Systemsandyljones.com
• Part 1: Key Concepts in RL - Spinning Up documentationspinningup.openai.com
• GitHub - heyseth/worm-sim: Simulate the C. Elegans worm brain in your browser and interact with the worm as it moves around · GitHubgithub.com
• Precise Manipulation with Efficient Online RLpi.website
• [2308.06578] The time is ripe to reverse engineer an entire nervous system: simulating behavior from neural interactionsarxiv.org
• Whole Brain Emulation: No Progress on C. elegans After 10 Years — LessWronglesswrong.com
• State of Robot Learning, December 2025vedder.io
• A Taxonomy of RL Environments for LLM Agentsleehanchung.github.io
• Deep Reinforcement Learning from Human Preferencesproceedings.neurips.cc
• UC Berkeley Library Proxy Loginwww-nature-com.libproxy.berkeley.edu
• RL Environments and RL for Science: Data Foundries and Multi-Agent Architecturesnewsletter.semianalysis.com