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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Matrix games like Prisoner's Dilemma have guided research on social dilemmasfor decades. However, they necessarily treat the choice to cooperate or defectas an atomic action. In real-world social dilemmas these choices are temporallyextended. Cooperativeness is a property that applies to policies, notelementary actions. We introduce sequential social dilemmas that share themixed incentive structure of matrix game social dilemmas but also requireagents to learn policies that implement their strategic intentions. We analyzethe dynamics of policies learned by multiple self-interested independentlearning agents, each using its own deep Q-network, on two Markov games weintroduce here: 1. a fruit Gathering game and 2. a Wolfpack hunting game. Wecharacterize how learned behavior in each domain changes as a function ofenvironmental factors including resource abundance. Our experiments show howconflict can emerge from competition over shared resources and shed light onhow the sequential nature of real world social dilemmas affects cooperation.

Multi-agent Reinforcement Learning in Sequential Social Dilemmas