A GPU-Parallel Framework for Heterogeneous Multi-Task Reinforcement Learning
A GPU-Parallel Framework for Heterogeneous Multi-Task Reinforcement Learning
Hebero 是一個基於 GPU 並行的 Isaac Lab 機械人學習基準,可在 40 項異質任務上聯合訓練及評估單一策略。研究提出 DGPO 及 IW-ABC;每項任務使用 50 個示範時,IW-ABC 的狀態輸入平均成功率達 90.1%,比最強基線 FAMO-ABC 高 7.8 個百分點,視覺版本則達 93.5%。
Published on Oct 6
Authors:
,
,
,
,
,
,
,
Abstract
GPU-parallel simulation provides abundant robot interaction, but existing benchmarks rarely combine this scale with heterogeneous manipulation tasks and standardized multi-task RL evaluation. We introduce Hebero (Heterogeneous Benchmark for Robot Learning), a GPU-parallel Isaac Lab benchmark that enables efficient joint training and evaluation of a single policy across all 40 heterogeneous tasks. Scaling experiments show that increasing parallel replicas per task improves success under a fixed wall-clock budget. To support learning with sparse rewards and limited demonstrations, we propose Demonstration-Guided Policy Optimization (DGPO), which reuses demonstrations for dense tracking rewards and asymmetric value learning. Its shared stack supports controlled comparisons of learner-specific demonstration interfaces within PPO. Within DGPO framework, we introduce IW-ABC, which uses a lightweight per-task learning progress signal to coordinate adaptive behavior cloning (ABC), relaxing demonstration guidance with task progress, and importance weighting (IW), emphasizing lagging tasks in PPO updates. With 50 demonstrations per task, IW-ABC achieves 90.1% state-input mean success, outperforming the strongest baseline FAMO-ABC by 7.8 percentage points. Its visual counterpart reaches 93.5% mean success. Real-world experiments further demonstrate that a single multi-task policy trained in simulation can successfully perform four tasks on a physical Piper robot. The project page is available at https://hebero-rl.github.io/.
View arXiv page View PDF Project page GitHub 2 Add to collection
Get this paper in your agent:
hf papers read 2606.03335
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash
Models citing this paper 0
No model linking this paper
Cite arxiv.org/abs/2606.03335 in a model README.md to link it from this page.
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2606.03335 in a dataset README.md to link it from this page.
Spaces citing this paper 0
No Space linking this paper
Cite arxiv.org/abs/2606.03335 in a Space README.md to link it from this page.
Collections including this paper 0
No Collection including this paper
Add this paper to a collection to link it from this page.
來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co