Metraipolis › archive › dc_21844ee702ab
Action Chunking Proximal Policy Optimization with Feedback Correction — arXiv:2609.36250
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.36250
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T17:28:53.584Z
- SHA-256 of the archived text
79902adcd7906c3a1c8bdbd754feca2910f01e1a9e0f7fa3696de12c58ced317- Stable URL of this page
https://metraipolis.com/documento/dc_21844ee702ab
How to cite
Action Chunking Proximal Policy Optimization with Feedback Correction — arXiv:2609.36250 — arXiv (en), https://arxiv.org/abs/2609.36250 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_21844ee702ab (Metraipolis archive dc_21844ee702ab, sha256 79902adcd7906c3a1c8bdbd754feca2910f01e1a9e0f7fa3696de12c58ced317)
BibTeX
@misc{dc21844ee702ab,
title = {Action Chunking Proximal Policy Optimization with Feedback Correction — arXiv:2609.36250},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.36250},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_21844ee702ab; sha256 79902adcd7906c3a1c8bdbd754feca2910f01e1a9e0f7fa3696de12c58ced317},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.36250. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Action chunking provides temporal abstraction in reinforcement learning by selecting short action sequences instead of individual actions, but many existing approaches face two limitations in high-dimensional robotic control. First, many rely on value functions over action chunks, which can be difficult to learn as action dimensionality and chunk length grow. Second, executing chunks open-loop removes within-chunk feedback, limiting reactivity in contact-rich tasks. We present Action Chunking PPO (ACPPO), a PPO extension that uses a chunked actor while retaining a standard state-value critic, thereby avoiding chunked Q-functions. We further propose ACPPO-Corr, which augments the chunk planner with a stepwise feedback corrector that adjusts planned actions online within each chunk. Across 25 simulated robotics tasks from IsaacGym and Bi-DexHands, spanning locomotion, arm manipulation, and dexterous hand-object interaction, ACPPO-Corr achieves the strongest aggregate performance among evaluated methods and performs best on both decision-frequency-sensitive and decision-frequency-neutral task subsets. Ablations show that moderate chunk lengths work best and that corrector regularization is important for balancing chunk-level planning with local feedback. These results suggest that action chunking can be effective in online PPO when chunk-level planning is paired with closed-loop correction. The code is available at: this https URL . Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.LG). Categorie arXiv: Machine Learning (cs.LG) ; Robotics (cs.RO). Pagina del lavoro: https://arxiv.org/abs/2609.36250 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-07T16:59:48.132Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).