AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Inferring the Future by Imagining the Past

Kartik Chandra, Tony Chen, Tzu-Mao Li, Jonathan Ragan-Kelley, Joshua B. Tenenbaum

NeurIPS 2023 · Probabilistic and scientific modeling › Molecules and proteins · approach: Counterfactual and statistical explanations · 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 backward-sampled importance sampling instead of rejection sampling (true posterior = Ours@1k)Posterior inference with 10 samples · Quantitative benchmarkBenchmark suite average TV: 0.063 -> 0.0257Improved
Use the proposed sampler instead of rejection sampling (true posterior = Rejection@10k)Posterior inference with 10 samples · Quantitative benchmarkKeys (observed holding no key) TV: 0.672 -> 0.410Improved
Allow uniform prior over start state instead of two fixed entrywaysProposed sampler · Quantitative benchmarkGrid TV vs Ours@1k: 0.159 -> 0.0538Improved
Proposed sampler versus rejection sampling on multi-stage keys task (pink key)Proposed sampler · Quantitative benchmarkKeys TV vs Rejection@10k: 0.456 -> 0.382Improved
Proposed sampler versus rejection sampling on green keyProposed sampler · Quantitative benchmarkKeys TV vs Rejection@10k: 0.463 -> 0.312Improved
Increase rejection sampling to 1,000 samples in blocks domainRejection sampling · Quantitative benchmarkValid samplesNo clear effect
Use rejection sampling with 10,000 samples as reference on BlocksRejection sampling · Quantitative benchmarkConvergenceHurt
Sample paths backward in time and only through the observed stateProposed sampler · Quantitative benchmarkQualitative posterior vs humans (N=200)Improved
Compare likelihoods of proposed sampler to rejection sampling at 25,000 samples each, with and without bidirectional path tracing (ablation)Proposed sampler · 4x4 gridworldLikelihood per cell, 4x4 gridworldNo clear effect
Apply sampler to continuous cart-pole with learned reverse dynamicsProposed sampler · Cart-poleVelocity posteriorImproved

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