Metraipolis › archive › dc_fa8a196da9c2

Neurosymbolic Reasoning with Incremental Knowledge for Sample Efficient Hierarchical Reinforcement Learning — arXiv:2608.02993

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.02993
License (as archived): CC0-1.0
Archived: 2026-10-31T12:15:29.161Z
SHA-256 of the archived text: 2c6f6e27cd20992db35141a659cbd42ccbae09302ade4ff4405e72b208e7bed6
Stable URL of this page: https://metraipolis.com/documento/dc_fa8a196da9c2

How to cite

Neurosymbolic Reasoning with Incremental Knowledge for Sample Efficient Hierarchical Reinforcement Learning — arXiv:2608.02993 — arXiv (en), https://arxiv.org/abs/2608.02993 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fa8a196da9c2 (Metraipolis archive dc_fa8a196da9c2, sha256 2c6f6e27cd20992db35141a659cbd42ccbae09302ade4ff4405e72b208e7bed6)

BibTeX

@misc{dcfa8a196da9c2,
  title = {Neurosymbolic Reasoning with Incremental Knowledge for Sample Efficient Hierarchical Reinforcement Learning — arXiv:2608.02993},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.02993},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fa8a196da9c2; sha256 2c6f6e27cd20992db35141a659cbd42ccbae09302ade4ff4405e72b208e7bed6},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.02993. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

(Flat) Reinforcement Learning (RL) agents face significant challenges in environments with sparse rewards that require long-horizon reasoning. A compelling approach to improve sample efficiency is to incorporate knowledge into learning and decision-making. In standard Hierarchical RL (HRL), knowledge is encoded in a fixed, non-updatable form, such as architectural choices, and remains unchanged throughout learning. With fixed HRL, reasoning with incremental knowledge learned during exploration is impractical before sufficient environmental knowledge is acquired, leading to poor sample efficiency. In this work, we propose neurosymbolic HRL with {\em Incremental Knowledge (InK)}: symbolic high-level components perform {\em symbolic planning} (e.g. using $D^*$) on an updatable representation of current InK, while low-level goal-conditioned neural modules learn motion primitives through experience using reward shaping. Experiments on navigation tasks demonstrate that incorporating InK substantially improves sample efficiency. Additionally, to perform {\em optimal} symbolic planning given {\em prior} knowledge about the world, we develop Belief World Tree Search. The code is available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-05 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2608.02993 — 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 2026-10-31T12:05:52.737Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).