Salvato in:
Dettagli Bibliografici
Autori principali: Jo, Yonghyeon, Lee, Sunwoo, Han, Seungyul
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2602.17062
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911699458588672
author Jo, Yonghyeon
Lee, Sunwoo
Han, Seungyul
author_facet Jo, Yonghyeon
Lee, Sunwoo
Han, Seungyul
contents Value decomposition is a core approach for cooperative multi-agent reinforcement learning (MARL). However, existing methods still rely on a single optimal action and struggle to adapt when the underlying value function shifts during training, often converging to suboptimal policies. To address this limitation, we propose Successive Sub-value Q-learning (S2Q), which learns multiple sub-value functions to retain alternative high-value actions. Incorporating these sub-value functions into a Softmax-based behavior policy, S2Q encourages persistent exploration and enables $Q^{\text{tot}}$ to adjust quickly to the changing optima. Experiments on challenging MARL benchmarks confirm that S2Q consistently outperforms various MARL algorithms, demonstrating improved adaptability and overall performance. Our code is available at https://github.com/hyeon1996/S2Q.
format Preprint
id arxiv_https___arxiv_org_abs_2602_17062
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Retaining Suboptimal Actions to Follow Shifting Optima in Multi-Agent Reinforcement Learning
Jo, Yonghyeon
Lee, Sunwoo
Han, Seungyul
Artificial Intelligence
Value decomposition is a core approach for cooperative multi-agent reinforcement learning (MARL). However, existing methods still rely on a single optimal action and struggle to adapt when the underlying value function shifts during training, often converging to suboptimal policies. To address this limitation, we propose Successive Sub-value Q-learning (S2Q), which learns multiple sub-value functions to retain alternative high-value actions. Incorporating these sub-value functions into a Softmax-based behavior policy, S2Q encourages persistent exploration and enables $Q^{\text{tot}}$ to adjust quickly to the changing optima. Experiments on challenging MARL benchmarks confirm that S2Q consistently outperforms various MARL algorithms, demonstrating improved adaptability and overall performance. Our code is available at https://github.com/hyeon1996/S2Q.
title Retaining Suboptimal Actions to Follow Shifting Optima in Multi-Agent Reinforcement Learning
topic Artificial Intelligence
url https://arxiv.org/abs/2602.17062