On Testing of Samplers

TL;DR

This paper introduces Barbarik2, a scalable testing algorithm that verifies whether a sampler's output distribution is close to a target distribution, effectively handling arbitrary weight functions with fewer samples than previous methods.

Contribution

Barbarik2 extends prior work by providing a rigorous, efficient testing method for general weighted samplers, overcoming limitations of uniform sampling verification.

Findings

Barbarik2 requires significantly fewer samples than previous methods.

It successfully tests state-of-the-art samplers with arbitrary weights.

The prototype implementation demonstrates practical applicability.

Abstract

Given a set of items $\mathcal{F}$ and a weight function $\mathtt{wt}: \mathcal{F} \mapsto (0,1)$, the problem of sampling seeks to sample an item proportional to its weight. Sampling is a fundamental problem in machine learning. The daunting computational complexity of sampling with formal guarantees leads designers to propose heuristics-based techniques for which no rigorous theoretical analysis exists to quantify the quality of generated distributions. This poses a challenge in designing a testing methodology to test whether a sampler under test generates samples according to a given distribution. Only recently, Chakraborty and Meel (2019) designed the first scalable verifier, called Barbarik1, for samplers in the special case when the weight function $\mathtt{wt}$ is constant, that is, when the sampler is supposed to sample uniformly from $\mathcal{F}$ . The techniques in Barbarik1, however, fail to handle general weight functions. The primary contribution of this paper is an affirmative answer to the above challenge: motivated by Barbarik1 but using different techniques and analysis, we design Barbarik2 an algorithm to test whether the distribution generated by a sampler is $\varepsilon$-close or $\eta$-far from any target distribution. In contrast to black-box sampling techniques that require a number of samples proportional to $|\mathcal{F}|$ , Barbarik2 requires only $\tilde{O}(tilt(\mathtt{wt},\varphi)^2/\eta(\eta - 6\varepsilon)^3)$ samples, where the $tilt$ is the maximum ratio of weights of two satisfying assignments. Barbarik2 can handle any arbitrary weight function. We present a prototype implementation of Barbarik2 and use it to test three state-of-the-art samplers.