Metraipolis › archive › dc_03f6980e2d75
CANDO: Conservative Action-Noise Directional Optimization for Residual Reinforcement Learning — arXiv:2611.04543
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.04543
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-04T12:29:57.619Z
- SHA-256 of the archived text
790ca1632dd12e194431c3372147b122abe3ee24ff55efd195f9a6429ec9c234- Stable URL of this page
https://metraipolis.com/documento/dc_03f6980e2d75
How to cite
CANDO: Conservative Action-Noise Directional Optimization for Residual Reinforcement Learning — arXiv:2611.04543 — arXiv (en), https://arxiv.org/abs/2611.04543 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_03f6980e2d75 (Metraipolis archive dc_03f6980e2d75, sha256 790ca1632dd12e194431c3372147b122abe3ee24ff55efd195f9a6429ec9c234)
BibTeX
@misc{dc03f6980e2d75,
title = {CANDO: Conservative Action-Noise Directional Optimization for Residual Reinforcement Learning — arXiv:2611.04543},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.04543},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_03f6980e2d75; sha256 790ca1632dd12e194431c3372147b122abe3ee24ff55efd195f9a6429ec9c234},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04543. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Pretrained generative policies provide strong behavior priors for robot control, but adapting them efficiently with online reinforcement learning remains difficult. A central challenge is how to improve a frozen generative policy without destroying the structure learned from offline demonstrations. Noise-space residual methods address this by modifying the policy's input noise, but existing noise-space methods typically regulate adaptation through fixed latent feasible regions, which may exclude valuable noise edits. In contrast, action-space residual methods are more sample-efficient, but their efficiency can still be further improved. We propose Conservative Noise Directional Optimization (CNDO), a noise-space residual RL method that improves a frozen generative policy by learning state- and noise-conditioned directions around Gaussian latent samples. CNDO removes predefined latent feasible regions and instead regulates noise-space adaptation through the soft conservative constraint, which consists of conservative value estimation and gated execution. We further introduce Conservative Action-Noise Directional Optimization (CANDO), which adds an action-space residual, combining structured latent-space improvement with local action correction. Experiments show that, once CNDO narrows the gap between noise-space and action-space residual learning, CANDO further improves over single-space optimization, achieving 76.1 NAUC and a 91.4% final success rate, surpassing the best single-space results by 6.6 NAUC and 4.1% success rate. More demos and info are available on this https URL . Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2611.04543 — 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 2026-11-04T12:26:47.281Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).