Two-sample testing

In statistical hypothesis testing, a two-sample test is a test performed on the data of two random samples, each independently obtained from a different given population. The purpose of the test is to determine whether the difference between these two populations is statistically significant. The statistics used in two-sample tests can be used to solve many machine learning problems, such as domain adaptation, covariate shift and generative adversarial networks.

Papers

Recently Added Most Hyped Most Active Needs Verification Most Verified

Showing 11–20 of 338 papers

Title	Date	Tasks	Status	Hype	Score
Towards Probabilistic Verification of Machine Unlearning	Mar 9, 2020	backdoor defenseMachine Unlearning	CodeCode Available	1	5
Addressing Maximization Bias in Reinforcement Learning with Two-Sample Testing	Jan 20, 2022	Q-Learningreinforcement-learning	CodeCode Available	1	5
Expert-Supervised Reinforcement Learning for Offline Policy Learning and Evaluation	Jun 23, 2020	reinforcement-learningReinforcement Learning	CodeCode Available	1	5
AI Feynman 2.0: Pareto-optimal symbolic regression exploiting graph modularity	Jun 18, 2020	regressionSymbolic Regression	CodeCode Available	1	5
Comparing distributions: _1 geometry improves kernel two-sample testing	Sep 19, 2019	Two-sample testingVocal Bursts Valence Prediction	CodeCode Available	0	5
Classification Logit Two-sample Testing by Neural Networks	Sep 25, 2019	ClassificationGeneral Classification	CodeCode Available	0	5
Comparing distributions: _1 geometry improves kernel two-sample testing	Dec 1, 2019	Two-sample testingVocal Bursts Valence Prediction	CodeCode Available	0	5
Adversarial Sample Detection for Deep Neural Network through Model Mutation Testing	Dec 14, 2018	SensitivityTwo-sample testing	CodeCode Available	0	5
Breaking hypothesis testing for failure rates	Jan 13, 2020	Point ProcessesTwo-sample testing	CodeCode Available	0	5
B-tests: Low Variance Kernel Two-Sample Tests	Jul 8, 2013	Two-sample testingVocal Bursts Valence Prediction	CodeCode Available	0	5

Show:10 25 50

← PrevPage 2 of 34Next →

All datasets Blob (9 modes, 40 for each)CIFAR-10 vs CIFAR-10.1 (1000 samples)HDGM (d=10, N=4000)HIGGS Data Set MNIST vs Fake MNIST

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	MMD-D	Avg accuracy	98.5	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	MMD-D	Avg accuracy	74.4	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	MMD-D	Avg accuracy	65.9	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	MMD-D	Avg accuracy	57.9	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	MMD-D	Avg accuracy	91	—	Unverified