AI Paper Atlas

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

Test-time Collective Prediction

Celestine Mendler-Dünner, Wenshuo Guo, Stephen Bates, Michael Jordan

NeurIPS 2021 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Interactive multi-agent trajectory prediction · 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 DeGroot adaptive weighting instead of equal-weight model averagingensemble of linear models · Synthetic logisticMSE: over 50x reductionImproved
use DeGroot consensus instead of picking best individual modelensemble · Synthetic logisticMSE: 5e^{-2} -> 4.6e^{-4}Improved
replace iterative power pooling with averaging trust scores (tau-avg) (ablation)DeGroot · Synthetic logisticMSE: 4.6e^{-4} -> 9e^{-3}Hurt
replace DeGroot weights with inverse of dataset-averaged MSE (ablation)DeGroot · Synthetic logisticMSE: 4.6e^{-4} -> 1e^{-2}Hurt
increase covariance of local feature distributions (more overlap) (ablation)DeGroot vs M-avg · Synthetic logisticgain over M-avgHurt
make local data variance extremely small (ablation)DeGroot · Synthetic logisticgain over M-avgHurt
choose neighbor count N too large (ablation)DeGroot · Synthetic logisticMSEHurt
choose very small N under highly noisy labels (ablation)DeGroot · Synthetic logisticMSEHurt
vary N between 1% and 10% of local data (ablation)DeGroot · Synthetic logisticMSENo clear effect
increase label/feature/model heterogeneity across agents (ablation)DeGroot vs M-avg · Abalone lassorelative gain over M-avgImproved
DeGroot vs M-avg on benchmarks (no validation data)DeGroot · Boston Ridgegain relative to DeGroot [%]: -12.45Improved
DeGroot vs CV-adaptive that uses extra validation dataDeGroot · E2006, Abalone, cpusmall, YearPredictiongain relative to DeGroot [%]: -0.39 +/- 0.75No clear effect

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