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

When solving a complex task, humans will spontaneously form teams and tocomplete different parts of the whole task, respectively. Meanwhile, thecooperation between teammates will improve efficiency. However, for currentcooperative MARL methods, the cooperation team is constructed through eitherheuristics or end-to-end blackbox optimization. In order to improve theefficiency of cooperation and exploration, we propose a structureddiversification emergence MARL framework named {\sc{Rochico}} based onreinforced organization control and hierarchical consensus learning.{\sc{Rochico}} first learns an adaptive grouping policy through theorganization control module, which is established by independent multi-agentreinforcement learning. Further, the hierarchical consensus module based on thehierarchical intentions with consensus constraint is introduced after teamformation. Simultaneously, utilizing the hierarchical consensus module and aself-supervised intrinsic reward enhanced decision module, the proposedcooperative MARL algorithm {\sc{Rochico}} can output the final diversifiedmulti-agent cooperative policy. All three modules are organically combined topromote the structured diversification emergence. Comparative experiments onfour large-scale cooperation tasks show that {\sc{Rochico}} is significantlybetter than the current SOTA algorithms in terms of exploration efficiency andcooperation strength.

Structured Diversification Emergence via Reinforced Organization Control and Hierarchical Consensus Learning