OpenAI Blog
May 16, 2017
Robots that learn
We’ve created a robotics system, trained entirely in simulation and deployed on a physical robot, which can learn a new task after seeing it done once.
Meta-learning for wrestling
We show that for the task of simulated robot wrestling, a meta-learning agent can learn to quickly defeat a stronger non-meta-learning agent, and also show that the meta-learning agent can adapt to physical malfunction.
Related stories
arXiv AI
Jul 22
The Open Ant: A Robot Platform for Reinforcement Learning Research
arXiv:2607. 18488v1 Announce Type: cross Abstract: Reinforcement learning (RL) research has demonstrated success in both physical and simulated domains; however, the predominant methodology remains rooted in simulations.
arXiv AI
Jun 19
Playful Agentic Robot Learning
arXiv:2606. 19419v1 Announce Type: cross Abstract: Current agentic robot systems can write executable Code-as-Policy programs, observe feedback, and revise behavior across multiple attempts, but they remain largely task-driven: reusable skills are acquired only after explicit instructions.
arXiv Machine Learning
Jun 8
Robotic Policy Adaptation via Weight-Space Meta-Learning
arXiv:2606. 07217v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models are emerging as a promising paradigm for robotic manipulation, enabling general-purpose policies trained from large corpora of demonstrations and action labels.
arXiv AI
Jul 2
Coachable agents for interactive gameplay
arXiv:2607. 00642v1 Announce Type: new Abstract: Reinforcement learning has proven to be a valuable tool in the creation of advanced AI and robotic systems, contributing to everything from game playing to robotics to foundation models.
arXiv AI
Jul 24
What Matters for Simulation to Online Reinforcement Learning on Real Robots
arXiv:2602. 20220v2 Announce Type: replace-cross Abstract: We investigate what specific design choices enable successful online reinforcement learning (RL) on physical robots.
arXiv AI
Jun 19
Superhuman Safe and Agile Racing through Multi-Agent Reinforcement Learning
arXiv:2605. 22748v2 Announce Type: replace-cross Abstract: Autonomous systems have achieved superhuman performance in isolation or simulation, yet they remain brittle in shared, dynamic real-world spaces.
arXiv AI
Jul 23
You Live More Than Once: Towards Hierarchical Skill Meta-Evolving
arXiv:2605. 28390v2 Announce Type: replace Abstract: Test-time skill evolving is regarded as a new paradigm for enhancing deployed agentic systems.
arXiv AI
1d ago
Teach and Grow: An Agent-Centered Architecture for General Robot Learning
arXiv:2608. 17209v1 Announce Type: cross Abstract: End-to-end vision-language-action (VLA) and world-action models offer an elegant route to general-purpose robotics, but their reliability is bounded by validated physical coverage.
OpenAI Blog
Apr 18, 2018
Evolved Policy Gradients
We’re releasing an experimental metalearning approach called Evolved Policy Gradients, a method that evolves the loss function of learning agents, which can enable fast training on novel tasks. Agents trained with EPG can succeed at basic tasks at test time that were outside their training regime, like learning to navigate to an object on a different side of the room from where it was placed during training.
arXiv Machine Learning
Jun 16
Task-Error Residual Learning for Real-Robot Five-Ball Juggling
arXiv:2606. 16978v1 Announce Type: cross Abstract: For residual learning that refines existing behavior, sample efficiency depends on two things: how much information each rollout returns, and how efficiently the learner uses that information.
arXiv Machine Learning
Jun 24
Debate2Create: Robot Co-design via Multi-Agent LLM Debate
arXiv:2510. 25850v3 Announce Type: replace-cross Abstract: We introduce Debate2Create (D2C), a multi-agent LLM framework that formulates robot co-design as structured, iterative debate grounded in physics-based evaluation.