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

Population-based multi-agent reinforcement learning (PB-MARL) refers to theseries of methods nested with reinforcement learning (RL) algorithms, whichproduces a self-generated sequence of tasks arising from the coupled populationdynamics. By leveraging auto-curricula to induce a population of distinctemergent strategies, PB-MARL has achieved impressive success in tacklingmulti-agent tasks. Despite remarkable prior arts of distributed RL frameworks,PB-MARL poses new challenges for parallelizing the training frameworks due tothe additional complexity of multiple nested workloads between sampling,training and evaluation involved with heterogeneous policy interactions. Tosolve these problems, we present MALib, a scalable and efficient computingframework for PB-MARL. Our framework is comprised of three key components: (1)a centralized task dispatching model, which supports the self-generated tasksand scalable training with heterogeneous policy combinations; (2) a programmingarchitecture named Actor-Evaluator-Learner, which achieves high parallelism forboth training and sampling, and meets the evaluation requirement ofauto-curriculum learning; (3) a higher-level abstraction of MARL trainingparadigms, which enables efficient code reuse and flexible deployments ondifferent distributed computing paradigms. Experiments on a series of complextasks such as multi-agent Atari Games show that MALib achieves throughputhigher than 40K FPS on a single machine with $32$ CPU cores; 5x speedup thanRLlib and at least 3x speedup than OpenSpiel in multi-agent training tasks.MALib is publicly available at https://github.com/sjtu-marl/malib.

MALib: A Parallel Framework for Population-based Multi-agent Reinforcement Learning