AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Sequential Predictive Two-Sample and Independence Testing

Aleksandr Podkopaev, Aaditya Ramdas

NeurIPS 2023 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Likelihood-free and graphical model estimation · 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 sequential predictive classification instead of the sequential kernelized two-sample testKDEF · KDEFobservations required to reach perfect powerImproved
Use sequential predictive independence testing instead of SKITMNIST same-digit dependence · MNISTdata required to achieve perfect powerImproved
Use predictive independence testing instead of SKITSpherical model · Spherical modelpowerHurt
Use predictive independence testing instead of SKITHTDD model · HTDD modelpowerImproved
Use predictive independence testing instead of SKITLocalized-dependence synthetic data · Localized-dependence synthetic datapowerImproved
Use Seq-C-IT instead of Seq-R-ITGaussian linear model · Gaussian linear modelpowerImproved
Use SKIT instead of predictive testsGaussian linear model · Gaussian linear modelpowerImproved
Use adaptive betting fractions rather than a fixed likelihood-ratio-style betLearned-predictor sequential two-sample testing · Gaussian two-sample exampletest powerImproved
Use fixed likelihood-ratio betting rather than adaptive betting fractionsBayes-optimal predictor sequential two-sample testing · Gaussian two-sample exampletest powerImproved
Use direct margin maximization instead of minimizing cross-entropy or squared lossPredictive two-sample testing · Two-sample testingtest powerHurt
Use a one-observation minibatch for the unbalanced-class testSequential two-sample testing with unknown class proportions · Unbalanced-class two-sample testingwealth growthHurt

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