AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Human spatiotemporal pattern learning as probabilistic program synthesis

Tracey Mills, Joshua B. Tenenbaum, Samuel J Cheyette

NeurIPS 2023 · Probabilistic and scientific modeling › Molecules and proteins · approach: Probabilistic embeddings and relational inference · 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
replace Bayesian Ridge regression baseline with LoT program-synthesis modelLoT vs Ridge · Human 2D sequence predictioncorrelation with human per-sequence error: 0.23 -> 0.56Improved
replace GP-NC with compositional GP-SLGP model · Human 2D sequence predictioncorrelation with human per-sequence error: 0.48 -> 0.50Improved
replace GP-SL with structured LoT modelGP-SL · Human 2D sequence predictioncorrelation with human per-sequence error: 0.50 -> 0.56Improved
replace GP-NC with LoT modelGP-NC · Human 2D sequence predictiondelta AIC / log-likelihood: -31592 -> -21835Improved
replace GP-SL with LoT modelGP-SL · Human 2D sequence predictionlog-likelihood of human data: -28416 -> -21835Improved
replace Bayesian Ridge with GP-SL (compositional kernels)Ridge · Human 2D sequence predictionlog-likelihood of human data: -30187 -> -28416Improved
replace GP-NC with Ridge regressionGP-NC · Human 2D sequence predictionlog-likelihood of human data: -31592 -> -30187Improved
add structure to the hypothesis space (GP-NC to GP-SL)GP-NC · Human 2D sequence predictiondelta AIC: 19514 -> 13162Improved
more observed points (first to final timepoint)human participants · Human 2D sequence predictionmean relative error: 0.94 -> 0.21Improved
more observed points beyond timepoint 10GP-NC and Ridge models · Human 2D sequence predictionrelative error over timeNo clear effect
more observed pointsLoT and GP-SL models · Human 2D sequence predictionrelative error over timeImproved
sequence complexity rating (higher)human participants · Human 2D sequence predictionmean relative error: r = 0.77Hurt

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