Meta Actor-Critic Framework for Multi-Agent Reinforcement Learning

Huang, Jiateng; Huang, Wanrong; Wu, Dan; Lan, Long

doi:10.1145/3488933.3489029

Cited by 1 publication

(2 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, even if explicit reward structures are crafted, agents might fall into reward manipulation during the policy learning and optimization phase (Skalse et al 2022). It involves exploiting vulnerabilities in the reward structure to maximize cumulative rewards without genuinely addressing the intended tasks (Leike et al 2018;Ouyang et al 2022).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Decoding Global Preferences: Temporal and Cooperative Dependency Modeling in Multi-Agent Preference-Based Reinforcement Learning

Zhu,

Qiu,

Zhou

et al. 2024

AAAI

View full text Add to dashboard Cite

Designing accurate reward functions for reinforcement learning (RL) has long been challenging. Preference-based RL (PbRL) offers a promising approach by using human preferences to train agents, eliminating the need for manual reward design. While successful in single-agent tasks, extending PbRL to complex multi-agent scenarios is nontrivial. Existing PbRL methods lack the capacity to comprehensively capture both temporal and cooperative aspects, leading to inadequate reward functions. This work introduces an advanced multi-agent preference learning framework that effectively addresses these limitations. Based on a cascading Transformer architecture, our approach captures both temporal and cooperative dependencies, alleviating issues related to reward uniformity and intricate interactions among agents. Experimental results demonstrate substantial performance improvements in multi-agent cooperative tasks, and the reconstructed reward function closely resembles expert-defined reward functions. The source code is available at https://github.com/catezi/MAPT.

show abstract

Section: Introductionmentioning

confidence: 99%

“…In summary, PbRL offers an effective way to acquire reward functions based on human intent rather than predefined designs. Its effectiveness spans diverse domains like robot control (Lee, Smith, and Abbeel 2021) and dialogue systems (Ouyang et al 2022).…”

Section: Introductionmentioning

confidence: 99%