Nearly Optimal Algorithms for Level Set Estimation

by   Blake Mason, et al.

The level set estimation problem seeks to find all points in a domain X where the value of an unknown function f: X→ℝ exceeds a threshold α. The estimation is based on noisy function evaluations that may be acquired at sequentially and adaptively chosen locations in X. The threshold value α can either be explicit and provided a priori, or implicit and defined relative to the optimal function value, i.e. α = (1-ϵ)f(x_∗) for a given ϵ > 0 where f(x_∗) is the maximal function value and is unknown. In this work we provide a new approach to the level set estimation problem by relating it to recent adaptive experimental design methods for linear bandits in the Reproducing Kernel Hilbert Space (RKHS) setting. We assume that f can be approximated by a function in the RKHS up to an unknown misspecification and provide novel algorithms for both the implicit and explicit cases in this setting with strong theoretical guarantees. Moreover, in the linear (kernel) setting, we show that our bounds are nearly optimal, namely, our upper bounds match existing lower bounds for threshold linear bandits. To our knowledge this work provides the first instance-dependent, non-asymptotic upper bounds on sample complexity of level-set estimation that match information theoretic lower bounds.


page 1

page 2

page 3

page 4


Sequential Experimental Design for Transductive Linear Bandits

In this paper we introduce the transductive linear bandit problem: given...

On the Optimal Bounds for Noisy Computing

We revisit the problem of computing with noisy information considered in...

High Dimensional Level Set Estimation with Bayesian Neural Network

Level Set Estimation (LSE) is an important problem with applications in ...

Settling the Sample Complexity of Single-parameter Revenue Maximization

This paper settles the sample complexity of single-parameter revenue max...

Lenient Regret and Good-Action Identification in Gaussian Process Bandits

In this paper, we study the problem of Gaussian process (GP) bandits und...

Learning Valuation Functions

In this paper we study the approximate learnability of valuations common...

Semi-Finite Length Analysis for Information Theoretic Tasks

We focus on the optimal value for various information-theoretical tasks....

Please sign up or login with your details

Forgot password? Click here to reset