arXiv research

The paper tackles pure exploration in multi-armed bandits with low rank structure using oblivious sampling.

This work introduces CAET, an algorithm for cost-aware pairwise pure exploration.

Unified algorithm for efficient pure exploration using dual variables.

Study on sample complexity for pure exploration in feedback graph settings.

Develops first optimal algorithm for logistic bandits.

Optimizes pure exploration in linear bandits with a new algorithm.

Study pure exploration in high-dimensional feature spaces using adaptive embeddings.

The study analyzes batched methods for early stopping in stochastic multi-armed bandits.

This paper tackles combinatorial pure exploration for dueling bandits, aiming to find the best candidate-position match.

This paper tackles combinatorial optimization under uncertainty with limited feedback.

UCB algorithm adapted for large-scale, non-sub-Gaussian problems.

Pure exploration (aka active testing) is the fundamental task of sequentially gathering information to answer a query about a stochastic environment. Good algorithms make few mistakes and take few samples. Lower bounds (for multi-armed bandit models with arms in an exponential family) reveal that the sample complexity …

We determine the sample complexity of pure exploration bandit problems with multiple good answers. We derive a lower bound using a new game equilibrium argument. We show how continuity and convexity properties of single-answer problems ensures that the Track-and-Stop algorithm has asymptotically optimal sample complexi…

New exploration bonuses improve reinforcement learning efficiency.

We study the Combinatorial Pure Exploration problem with Continuous and Separable reward functions (CPE-CS) in the stochastic multi-armed bandit setting. In a CPE-CS instance, we are given several stochastic arms with unknown distributions, as well as a collection of possible decisions. Each decision has a reward accor…

We study a specific \textit{combinatorial pure exploration stochastic bandit problem} where the learner aims at finding the set of arms whose means are above a given threshold, up to a given precision, and \textit{for a fixed time horizon}. We propose a parameter-free algorithm based on an original heuristic, and prove…

The paper tackles combinatorial pure exploration with various feedback structures and proposes efficient algorithms.

In this paper, we consider the challenge of maximizing an unknown function f for which evaluations are noisy and are acquired with high cost. An iterative procedure uses the previous measures to actively select the next estimation of f which is predicted to be the most useful. We focus on the case where the function ca…

Algorithm identifies optimal stable matching in uncertain two-sided markets.

New algorithm for collaborative bandit learning reduces sample complexity and regret.

We propose the first fully-adaptive algorithm for pure exploration in linear bandits---the task to find the arm with the largest expected reward, which depends on an unknown parameter linearly. While existing methods partially or entirely fix sequences of arm selections before observing rewards, our method adaptively c…

New algorithm finds high-reward combinatorial sets with fewest pulls.

New methods for efficient exploration under unknown linear constraints in bandits.

New algorithms find all ε-good arms in stochastic bandits.

New algorithm identifies optimal actions in large reward spaces efficiently.

Improved batched SH algorithm maintains original performance.

New algorithms optimize decision-making under uncertainty with graph information.

New algorithm explores optimally using same primitives as Thompson Sampling.

New algorithms for model selection in linear bandits adapt to instance complexity.

Optimal algorithm for selecting high-quality arms from infinite bandit arms.

Efficiently clusters noisy data with minimal queries.

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, whi…

In several realistic situations, an interactive learning agent can practice and refine its strategy before going on to be evaluated. For instance, consider a student preparing for a series of tests. She would typically take a few practice tests to know which areas she needs to improve upon. Based of the scores she obta…

We study the combinatorial pure exploration problem Best-Set in stochastic multi-armed bandits. In a Best-Set instance, we are given $n$ arms with unknown reward distributions, as well as a family $\mathcal{F}$ of feasible subsets over the arms. Our goal is to identify the feasible subset in $\mathcal{F}$ with the maxi…

Study on identifying most preferred policy in bandits with vector-valued rewards.

Improved confidence bounds for linear logistic model with applications to bandits.

There are two variants of the classical multi-armed bandit (MAB) problem that have received considerable attention from machine learning researchers in recent years: contextual bandits and simple regret minimization. Contextual bandits are a sub-class of MABs where, at every time step, the learner has access to side in…

New algorithm achieves near optimal sample complexity for 1-identification problem.

Unified framework for ranking-and-selection with multiple correct answers and non-answerable estimates

Optimal algorithm for identifying best arm in stochastic linear bandits with fixed confidence.

New algorithm identifies best policy in MDPs faster.

Algorithm identifies best arm in combinatorial bandits with semi-bandit feedback.

Paper develops a dynamic Bayesian approach for active learning that optimizes exploration-exploitation balance.

We study the problem of stochastic combinatorial pure exploration (CPE), where an agent sequentially pulls a set of single arms (a.k.a. a super arm) and tries to find the best super arm. Among a variety of problem settings of the CPE, we focus on the full-bandit setting, where we cannot observe the reward of each singl…

FraPPE efficiently identifies Pareto optimal arms in multi-objective bandits.

This work optimizes identifying good arms in nonparametric multi-armed bandits.

New experimental design minimizes regret in bandits.

Algorithm optimizes a single attribute in multi-armed bandits with constraints.