Sequential algorithms for testing identity and closeness of distributions

by   Omar Fawzi, et al.

What advantage do sequential procedures provide over batch algorithms for testing properties of unknown distributions? Focusing on the problem of testing whether two distributions π’Ÿ_1 and π’Ÿ_2 on {1,…, n} are equal or Ο΅-far, we give several answers to this question. We show that for a small alphabet size n, there is a sequential algorithm that outperforms any batch algorithm by a factor of at least 4 in terms sample complexity. For a general alphabet size n, we give a sequential algorithm that uses no more samples than its batch counterpart, and possibly fewer if the actual distance TV(π’Ÿ_1, π’Ÿ_2) between π’Ÿ_1 and π’Ÿ_2 is larger than Ο΅. As a corollary, letting Ο΅ go to 0, we obtain a sequential algorithm for testing closeness when no a priori bound on TV(π’Ÿ_1, π’Ÿ_2) is given that has a sample complexity π’ͺΜƒ(n^2/3/TV(π’Ÿ_1, π’Ÿ_2)^4/3): this improves over the π’ͺΜƒ(n/log n/TV(π’Ÿ_1, π’Ÿ_2)^2) tester of <cit.> and is optimal up to multiplicative constants. We also establish limitations of sequential algorithms for the problem of testing identity and closeness: they can improve the worst case number of samples by at most a constant factor.

βˆ™ 10/28/2017

Wasserstein Identity Testing

Uniformity testing and the more general identity testing are well studie...
βˆ™ 11/17/2019

Testing Properties of Multiple Distributions with Few Samples

We propose a new setting for testing properties of distributions while r...
βˆ™ 06/21/2022

Sharp Constants in Uniformity Testing via the Huber Statistic

Uniformity testing is one of the most well-studied problems in property ...
βˆ™ 01/17/2018

An Empirical Analysis of Proximal Policy Optimization with Kronecker-factored Natural Gradients

In this technical report, we consider an approach that combines the PPO ...
βˆ™ 06/24/2023

On Scalable Testing of Samplers

In this paper we study the problem of testing of constrained samplers ov...
βˆ™ 09/06/2023

Testing properties of distributions in the streaming model

We study distribution testing in the standard access model and the condi...
βˆ™ 02/10/2019

The Optimal Approximation Factor in Density Estimation

Consider the following problem: given two arbitrary densities q_1,q_2 an...

Please sign up or login with your details

Forgot password? Click here to reset