DNF Learning via Locally Mixing Random Walks

TL;DR

This paper introduces a quasi-polynomial time algorithm for PAC learning DNF formulas with membership queries in a distribution-free setting, utilizing locally mixing random walks to identify terms.

Contribution

It presents the first quasi-polynomial time list-decoding algorithm for learning a single term of a DNF and applies it to learn size-s DNFs with all terms of the same size.

Findings

Algorithm successfully identifies candidate terms with high probability.

New analysis of locally mixing random walks enables efficient learning.

Approach works under arbitrary unknown distributions.

Abstract

We give two results on PAC learning DNF formulas using membership queries in the challenging "distribution-free" learning framework, where learning algorithms must succeed for an arbitrary and unknown distribution over $\{0,1\}^n$. (1) We first give a quasi-polynomial time "list-decoding" algorithm for learning a single term of an unknown DNF formula. More precisely, for any target $s$-term DNF formula $f = T_1 \vee \cdots \vee T_s$ over $\{0,1\}^n$ and any unknown distribution $D$ over $\{0,1\}^n$, our algorithm, which uses membership queries and random examples from $D$, runs in $\textsf{quasipoly}(n,s)$ time and outputs a list $L$ of candidate terms such that with high probability some term $T_i$ of $f$ belongs to $L$. (2) We then use result (1) to give a $\textsf{quasipoly}(n,s)$-time algorithm, in the distribution-free PAC learning model with membership queries, for learning the class of size-$s$ DNFs in which all terms have the same size. Our algorithm learns using a DNF hypothesis. The key tool used to establish result (1) is a new result on "locally mixing random walks," which, roughly speaking, shows that a random walk on a graph that is covered by a small number of expanders has a non-negligible probability of mixing quickly in a subset of these expanders.