SOTAVerified

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 AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 181–190 of 338 papers

TitleStatusHype
Adversarial Sample Detection for Deep Neural Network through Model Mutation TestingCode0
Sequential Experiment Design for Hypothesis Verification—0
Testing for Families of Distributions via the Fourier Transform—0
Practical methods for graph two-sample testingCode0
Testing Changes in Communities for the Stochastic Block Model—0
The Structure of Optimal Private Tests for Simple Hypotheses—0
Variational Autoencoders for New Physics Mining at the Large Hadron ColliderCode0
Understanding Learned Models by Identifying Important Features at the Right ResolutionCode0
Minimax Rates in Network Analysis: Graphon Estimation, Community Detection and Hypothesis Testing—0
How Secure are Deep Learning Algorithms from Side-Channel based Reverse Engineering?—0
Show:102550
← PrevPage 19 of 34Next →
All datasetsBlob (9 modes, 40 for each)CIFAR-10 vs CIFAR-10.1 (1000 samples)HDGM (d=10, N=4000)HIGGS Data SetMNIST vs Fake MNIST

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1MMD-DAvg accuracy98.5—Unverified
#ModelMetricClaimedVerifiedStatus
1MMD-DAvg accuracy74.4—Unverified
#ModelMetricClaimedVerifiedStatus
1MMD-DAvg accuracy65.9—Unverified
#ModelMetricClaimedVerifiedStatus
1MMD-DAvg accuracy57.9—Unverified
#ModelMetricClaimedVerifiedStatus
1MMD-DAvg accuracy91—Unverified