AI Paper Atlas

Research map / Multi-agent systems and games / Human-AI and goal-conditioned

Beyond Rewards: a Hierarchical Perspective on Offline Multiagent Behavioral Analysis

Shayegan Omidshafiei, Andrei Kapishnikov, Yannick Assogba, Lucas Dixon, Been Kim

NeurIPS 2022 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: World-model-based agents · 1 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 hierarchical joint+local latents (MOHBA) instead of an LSTM final-hidden-state encodingMOHBA · Hill-climbing domainAPL (lower better): 4.8 -> 1.3Improved
use hierarchical joint+local latents (MOHBA) instead of an LSTM final-hidden-state encodingMOHBA · Coordination gameAPL (lower better): 2.9 -> 2.4Improved
use hierarchical latents instead of flat-VAE feeding only z_omega to policies (remove local latents and local KL) (ablation)MOHBA · Hill-climbing domainAPL (lower better): 6.9 -> 1.3Improved
use hierarchical latents instead of flat-VAE feeding only z_omega to policies (ablation)MOHBA · Coordination gameAPL (lower better): 5.0 -> 2.4Improved
use MOHBA joint latent instead of LSTM hidden state for K-means clusteringMOHBA · Hill-climbing domainICTD (lower better): 0.47 -> 0.17Improved
use MOHBA joint latent instead of LSTM hidden state for K-means clusteringMOHBA · Coordination gameICTD (lower better): 0.34 -> 0.16Improved
add hierarchy (local latents) over flat-VAE for joint clustering quality (ablation)MOHBA · Hill-climbing domainICTD (lower better): 0.18 -> 0.17No clear effect
add hierarchy (local latents) over flat-VAE for joint clustering quality (ablation)MOHBA · Coordination gameICTD (lower better): 0.16 -> 0.16No clear effect
flat-VAE baseline still recovers the 3 coordination clusters, whereas LSTM hidden states do notLSTM vs VAE baseline · Coordination gamelatent cluster interpretability (qualitative)Mixed
predict dispersion class from z_omega only (instead of full trajectory)MOHBA · Hill-climbing agent dispersion (5 classes)validation accuracy: 54.27%Improved
predict total-return class from z_omega onlyMOHBA · Hill-climbing total return (5 classes)validation accuracy: 60.98%Improved
cyclically anneal the KL weight beta from 0 to target valueMOHBA training · training stabilityKL vanishing (qualitative)Improved

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