AI Paper Atlas

Research map / Multi-agent systems and games / Cooperative multi-agent RL

Explaining Decentralized Multi-Agent Reinforcement Learning Policies

Kayla Boggess, Sarit Kraus, Lu Feng

AAAI 2026 · Multi-agent systems and games › Cooperative multi-agent RL · approach: Cooperative MARL algorithms · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use Hasse diagram summarization instead of per-agent abstract policy graphsdecentralized MARL policy summary · SR (9,7) CTDEnumber of nodes: 534 -> 8Improved
use Hasse diagram summarization instead of per-agent abstract policy graphsdecentralized MARL policy summary · RW (4,19) CTDEnumber of nodes: 1,274 -> 20Improved
use Hasse diagram summarization instead of per-agent abstract policy graphsdecentralized MARL policy summary · PP (7,6) DTDEnumber of nodes: 107 -> 7Improved
show HDS summaries to users instead of baseline graphsuser question answering · Summarization user studycorrect answers out of 6: 3.1 -> 4.25Improved
use HDS instead of baselineuser-rated completeness · Summarization user studyLikert completeness: W=16.0, Z=-2.07, p <= 0.04Improved
use HDS instead of baselineuser-rated summary quality on other dimensions · Summarization user studyLikert ratingsNo clear effect
use HD-When explanations (with uncertain features) instead of union baselineexplanation size · SR (9,7) CTDEcertain features: 54 -> 9Improved
use HD-When instead of baselineexplanation size · RW (4,19) CTDEcertain features: 267 -> 0Mixed
use HD explanations instead of baseline for When queriesuser prediction accuracy · Explanation user studycorrect answers (when): t(20)=9.65, d=2.16Improved
use HD explanations instead of baseline for Why Not queriesuser prediction accuracy · Explanation user studycorrect answers (why not): t(20)=13.23, d=2.96Improved
use HD explanations instead of baseline for What queriesuser prediction accuracy · Explanation user studycorrect answers (what): t(20)=12.05, d=2.69Improved
use HD explanations instead of baselineuser ratings of explanation quality · Explanation user studyLikert understanding: W=3.5, Z=-3.02, r=-0.47Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy