AI Paper Atlas

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

THOMAS: Trajectory Heatmap Output with learned Multi-Agent Sampling

Thomas Gilles, Stefano Sabatini, Dzmitry Tsishkou, Bogdan Stanciulescu, Fabien Moutarde

ICLR 2022 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Interactive multi-agent trajectory prediction · 57 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 learned scene-level trajectory recombination to marginal predictions (ablation)THOMAS · Interaction validationJoint MR: 14.8 -> 11.8Improved
add learned scene-level trajectory recombination to marginal predictions (ablation)THOMAS · Interaction validationCollision rate: 7.2% to 2.4%Improved
replace learned recombination with collision-aware algorithmic sampling (ablation)THOMAS · Interaction validationCollision rate: 7.2 -> 2.6Improved
replace learned recombination with collision-aware algorithmic sampling (ablation)THOMAS · Interaction validationJoint MR: 14.8 -> 14.8No clear effect
use learned sampling instead of deterministic sampling followed by recombination (ablation)THOMAS · Interaction validationJoint MR: 11.8 -> 15.2Hurt
replace heatmap output with scalar regression and a joint objective (ablation)THOMAS · Interaction validationMarginal MR: 8.2 -> 16.2Hurt
use marginal rather than joint training objective with algorithmic sampling (ablation)THOMAS · Interaction validationCollision rate: 2.6 -> 7.2Hurt
use hierarchical heatmap refinement instead of GOHOME lane rasters (ablation)THOMAS · Inference with 128 agentsinference time: 90 ms -> 31 msImproved
use hierarchical heatmap refinement instead of GOHOME lane rasters (ablation)THOMAS · Interaction validationmarginal MissRate6No clear effect
use THOMAS instead of SceneTransformerjoint multi-agent prediction · Interaction validationJoint MR: 15.7 -> 11.8Improved
use THOMAS instead of DenseTNTjoint multi-agent prediction · Interaction regular scene testcSMR: 22.4 -> 25.2Mixed

Later papers that compared with this method

3 later accepted papers report THOMAS as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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