Co-design Gym: A Unified Benchmark for Embodiment-Policy Co-optimization

arXiv:2610.02366 · cs.LG, cs.RO · Submitted 2026-10-01 · Read on arXiv

cs.LG, cs.RO

Submitted: 2026-10-01

Updated: 2026-10-01

Comments: Aviraj Newatia and Yordan Tsvetkov contributed equally

Code: https://github.com/camral/codesign-gym

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Benchmark — 1× in this paper · explained in 2 episodes

Sources

• Dota 2 with Large Scale Deep Reinforcement Learning
• OpenAI Gym
• A Mathematical Theory of Co-Design
• Transformer Transformer: A Unified Model for Motion-Conditioned Robot Co-design
• Task-Agnostic Morphology Evolution
• pymgrid: An Open-Source Python Microgrid Simulator for Applied Artificial Intelligence Research
• Learning to Jump from Pixels
• Large Language Models: A Survey
• Proximal Policy Optimization Algorithms
• FastTD3: Simple, Fast, and Capable Reinforcement Learning for Humanoid Control
• Efficient and Versatile Quadrupedal Skating: Optimal Co-design via Reinforcement Learning and Bayesian Optimization
• BRIDGE: An Open-Source Humanoid Platform via Morphology-Control Co-Design for Physical AI
• Neural Graph Evolution: Towards Efficient Automatic Robot Design
• Dynamics-Guided Diffusion Model for Sensor-less Robot Manipulator Design
• Transform2Act: Learning a Transform-and-Control Policy for Efficient Agent Design

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks