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

In cooperative multi-agent reinforcement learning, a team of agents workstogether to achieve a common goal. Different environments or tasks may requirevarying degrees of coordination among agents in order to achieve the goal in anoptimal way. The nature of coordination will depend on the properties of theenvironment -- its spatial layout, distribution of obstacles, dynamics, etc. Weterm this variation of properties within an environment as heterogeneity.Existing literature has not sufficiently addressed the fact that differentenvironments may have different levels of heterogeneity. We formalize thenotions of coordination level and heterogeneity level of an environment andpresent HECOGrid, a suite of multi-agent RL environments that facilitatesempirical evaluation of different MARL approaches across different levels ofcoordination and environmental heterogeneity by providing a quantitativecontrol over coordination and heterogeneity levels of the environment. Further,we propose a Centralized Training Decentralized Execution learning approachcalled Stateful Active Facilitator (SAF) that enables agents to workefficiently in high-coordination and high-heterogeneity environments through adifferentiable and shared knowledge source used during training and dynamicselection from a shared pool of policies. We evaluate SAF and compare itsperformance against baselines IPPO and MAPPO on HECOGrid. Our results show thatSAF consistently outperforms the baselines across different tasks and differentheterogeneity and coordination levels. We release the code for HECOGrid as wellas all our experiments.

Stateful active facilitator: Coordination and Environmental Heterogeneity in Cooperative Multi-Agent Reinforcement Learning