English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Multi-agent reinforcement learning is difficult to be applied in practice,which is partially due to the gap between the simulated and real-worldscenarios. One reason for the gap is that the simulated systems always assumethat the agents can work normally all the time, while in practice, one or moreagents may unexpectedly "crash" during the coordination process due toinevitable hardware or software failures. Such crashes will destroy thecooperation among agents, leading to performance degradation. In this work, wepresent a formal formulation of a cooperative multi-agent reinforcementlearning system with unexpected crashes. To enhance the robustness of thesystem to crashes, we propose a coach-assisted multi-agent reinforcementlearning framework, which introduces a virtual coach agent to adjust the crashrate during training. We design three coaching strategies and the re-samplingstrategy for our coach agent. To the best of our knowledge, this work is thefirst to study the unexpected crashes in the multi-agent system. Extensiveexperiments on grid-world and StarCraft II micromanagement tasks demonstratethe efficacy of adaptive strategy compared with the fixed crash rate strategyand curriculum learning strategy. The ablation study further illustrates theeffectiveness of our re-sampling strategy.

Coach-assisted Multi-Agent Reinforcement Learning Framework for Unexpected Crashed Agents