AI Paper Atlas

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

Relax, it doesn’t matter how you get there: A new self-supervised approach for multi-timescale behavior analysis

Mehdi Azabou, Michael Jacob Mendelson, Nauman Ahad, Maks Sorokin, Shantanu Thakoor, Carolina Urzay, Eva L Dyer

NeurIPS 2023 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Imitation and skill learning · 4 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
Add the Histogram of Actions prediction objective (ablation)BAMS · MABe Mouse TripletsAverage sequence-level MSE: 0.093100 -> 0.090415Improved
Add the Histogram of Actions prediction objective (ablation)BAMS · MABe Mouse TripletsAverage sequence-level F1: 68.30 -> 80.12Improved
Add the Histogram of Actions prediction objective (ablation)BAMS · MABe Mouse TripletsAverage frame-level F1: 19.46 -> 19.85Improved
Add multi-timescale bootstrapping (ablation)BAMS · MABe Mouse TripletsAverage sequence-level F1: 78.11 -> 80.12Improved
Add multi-timescale bootstrapping (ablation)BAMS · MABe Mouse TripletsAverage frame-level F1: 19.45 -> 19.85Improved
Separate the short- and long-term representations instead of using one space (ablation)BAMS · MABe Mouse TripletsAverage sequence-level F1: 73.37 -> 80.12Improved
Separate the short- and long-term representations instead of using one space (ablation)BAMS · MABe Mouse TripletsAverage frame-level F1: 19.52 -> 19.85Improved
Use temporal contrastive loss in TS2Vec rather than instance contrastive loss (ablation)TS2Vec · MABe Mouse Triplets — StrainF1-score: 57.12 -> 45.82Hurt
Use temporal contrastive loss in TS2Vec rather than instance contrastive loss (ablation)TS2Vec · MABe Mouse Triplets — HuddleF1-score: 24.74 -> 9.38Hurt
Pre-train on the full dataset rather than the training split onlyBAMS · MABe Mouse Triplets — StrainF1-score: 83.44 -> 88.23Improved
Use HoA future-distribution prediction instead of sequential future prediction (ablation)BAMS · MABe Mouse TripletsMaximum prediction horizon: 10 -> 30Improved

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