Metraipolis › archive › dc_ac77dcff0667
Arithmetic Actor Heads and Training Stabilization for Out-of-Distribution Reinforcement Learning — arXiv:2610.05143
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.05143
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T05:30:56.103Z
- SHA-256 of the archived text
18815990f0f3ea400175cc7e97afe785b4c81209b0d1b2d47accabd623ff5d8c- Stable URL of this page
https://metraipolis.com/documento/dc_ac77dcff0667
How to cite
Arithmetic Actor Heads and Training Stabilization for Out-of-Distribution Reinforcement Learning — arXiv:2610.05143 — arXiv, https://arxiv.org/abs/2610.05143 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ac77dcff0667 (Metraipolis archive dc_ac77dcff0667, sha256 18815990f0f3ea400175cc7e97afe785b4c81209b0d1b2d47accabd623ff5d8c)
BibTeX
@misc{dcac77dcff0667,
title = {Arithmetic Actor Heads and Training Stabilization for Out-of-Distribution Reinforcement Learning — arXiv:2610.05143},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.05143},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ac77dcff0667; sha256 18815990f0f3ea400175cc7e97afe785b4c81209b0d1b2d47accabd623ff5d8c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.05143. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Reinforcement learning (RL) policies can deteriorate under out-of-distribution (OOD) magnitude shifts. Starting from soft actor-critic (SAC) and its Bayesian Amnesic Piecewise-Robust (BAPR) predecessor, we study the causal-symbolic BAPR (CS-BAPR) family. The practical method combines six training-stabilization settings with alternative actor heads: a Neural Addition Unit (NAU) with a Neural Multiplication Unit (NMU)-inspired quadratic correction, a Kolmogorov-Arnold Network (KAN), or a multilayer perceptron (MLP) with rectified linear unit (ReLU) or hyperbolic-tangent activations. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2610.05143 — 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 2026-10-06T05:27:05.769Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).