analiza aiqbench
This paper introduces MT-Libero, a GPU-parallel benchmark for multi-task robot manipulation that enables simultaneous training across heterogeneous suites using parallel rendering and physics randomization. The proposed DGPO method improves practical training under sparse rewards by combining importance-weighted PPO with adaptive behavior cloning, offering a tunable balance between demonstration guidance and on-policy stability. For robotics researchers, this framework addresses the current bottleneck of single-task specialization while providing a scalable infrastructure for complex simulation environments.
GPU-Parallel Multi-Task Reinforcement Learning with Demonstration Guided Policy Optimization
arXiv:2606.03335v1 Announce Type: new Abstract: Large scale GPU-parallel reinforcement learning has changed what can be trained in robot simulation, yet most systems still optimize one specialist policy per task. We propose a construction methodology for turning structured manipulation task families into GPU-parallel multi-task RL benchmarks, and instantiate it as MT-Libero using LIBERO assets and task predicates in Isaac Lab. The resulting benchmark supports simultaneous reinforcement learning over heterogeneous task suites with parallel rendering, physics randomization, and state-input or visual-input policies. To make such training practical under sparse success signals and limited prior data, we further propose DGPO, an on-policy demonstration guided method that combines importance weighted PPO with adaptive behavior cloning on matched demonstration actions. DGPO enables a tunable preference toward demonstrated task distributions, outperforming both prior-free RL and existing demonstration-based methods while preserving the stability and online improvement benefits of on-policy PPO.
Abstract:Large scale GPU-parallel reinforcement learning has changed what can be trained in robot simulation, yet most systems still optimize one specialist policy per task. We propose a construction methodology for turning structured manipulation task families into GPU-parallel multi-task RL benchmarks, and instantiate it as MT-Libero using LIBERO assets and task predicates in Isaac Lab. The resulting benchmark supports simultaneous reinforcement learning over heterogeneous task suites with parallel rendering, physics randomization, and state-input or visual-input policies. To make such training practical under sparse success signals and limited prior data, we further propose DGPO, an on-policy demonstration guided method that combines importance weighted PPO with adaptive behavior cloning on matched demonstration actions. DGPO enables a tunable preference toward demonstrated task distributions, outperforming both prior-free RL and existing demonstration-based methods while preserving the stability and online improvement benefits of on-policy PPO.
Robotics (cs.RO)
arXiv:2606.03335 [cs.RO]
(or arXiv:2606.03335v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2606.03335 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 231 cuvinte
Subiecte conexe
[AINews] OpenAI reports Navier-Stokes singularity find in 88 hours using Astra-next, roughly 10,000 agents and 130B tokens (>$40M), a contender for second ever Millennium Prize awarded
Latent Space · 9 sept. 2026
A GPU-Parallel Framework for Heterogeneous Multi-Task Reinforcement Learning
arXiv · 6 oct. 2026
Mastering Agentic Techniques: AI Agent Reinforcement Learning
NVIDIA · 1 iul. 2026
A Balanced Data Diet: Addressing the Exploration Bottleneck in Mega-Scale RL for Robot Control
arXiv · 9 oct. 2026