pith. sign in

Disagreement-Based Combinatorial Pure Exploration: Sample Complexity Bounds and an Efficient Algorithm

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

We design new algorithms for the combinatorial pure exploration problem in the multi-arm bandit framework. In this problem, we are given $K$ distributions and a collection of subsets $\mathcal{V} \subset 2^{[K]}$ of these distributions, and we would like to find the subset $v \in \mathcal{V}$ that has largest mean, while collecting, in a sequential fashion, as few samples from the distributions as possible. In both the fixed budget and fixed confidence settings, our algorithms achieve new sample-complexity bounds that provide polynomial improvements on previous results in some settings. Via an information-theoretic lower bound, we show that no approach based on uniform sampling can improve on ours in any regime, yielding the first interactive algorithms for this problem with this basic property. Computationally, we show how to efficiently implement our fixed confidence algorithm whenever $\mathcal{V}$ supports efficient linear optimization. Our results involve precise concentration-of-measure arguments and a new algorithm for linear programming with exponentially many constraints.

fields

stat.ML 1

years

2019 1

verdicts

UNVERDICTED 1

representative citing papers

Sequential Experimental Design for Transductive Linear Bandits

stat.ML · 2019-06-20 · unverdicted · novelty 7.0

Introduces transductive linear bandits, gives instance-dependent lower bounds, and presents an algorithm matching them up to logarithmic factors, including the first non-asymptotic near-optimal method for standard linear bandits.

citing papers explorer

Showing 1 of 1 citing paper.

  • Sequential Experimental Design for Transductive Linear Bandits stat.ML · 2019-06-20 · unverdicted · none · ref 7 · internal anchor

    Introduces transductive linear bandits, gives instance-dependent lower bounds, and presents an algorithm matching them up to logarithmic factors, including the first non-asymptotic near-optimal method for standard linear bandits.