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/

Global: Zendy Plus annual

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

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Team competition in multi-agent Markov games is an increasingly importantsetting for multi-agent reinforcement learning, due to its generalapplicability in modeling many real-life situations. Multi-agent actor-criticmethods are the most suitable class of techniques for learning optimal policiesin the team competition setting, due to their flexibility in learningagent-specific critic functions, which can also learn from other agents. Inmany real-world team competitive scenarios, the roles of the agents naturallyemerge, in order to aid in coordination and collaboration within members of theteams. However, existing methods for learning emergent roles rely heavily onthe Q-learning setup which does not allow learning of agent-specificQ-functions. In this paper, we propose RAC, a novel technique for learning theemergent roles of agents within a team that are diverse and dynamic. In theproposed method, agents also benefit from predicting the roles of the agents inthe opponent team. RAC uses the actor-critic framework with role encoder andopponent role predictors for learning an optimal policy. Experimentation using2 games demonstrates that the policies learned by RAC achieve higher rewardsthan those learned using state-of-the-art baselines. Moreover, experimentssuggest that the agents in a team learn diverse and opponent-aware policies.

Opponent-aware Role-based Learning in Team Competitive Markov Games