Saved in:
Bibliographic Details
Main Authors: Ghosh, Ayon, Prashanth, L. A., Sen, Dipayan, Gopalan, Aditya
Format: Preprint
Published: 2022
Subjects:
Online Access:https://arxiv.org/abs/2203.16810
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910925011812352
author Ghosh, Ayon
Prashanth, L. A.
Sen, Dipayan
Gopalan, Aditya
author_facet Ghosh, Ayon
Prashanth, L. A.
Sen, Dipayan
Gopalan, Aditya
contents We consider the problem of sequentially learning to estimate, in the mean squared error (MSE) sense, a Gaussian $K$-vector of unknown covariance by observing only $m < K$ of its entries in each round. We propose two MSE estimators, and analyze their concentration properties. The first estimator is non-adaptive, as it is tied to a predetermined $m$-subset and lacks the flexibility to transition to alternative subsets. The second estimator, which is derived using a regression framework, is adaptive and exhibits better concentration bounds in comparison to the first estimator. We frame the MSE estimation problem with bandit feedback, where the objective is to find the MSE-optimal subset with high confidence. We propose a variant of the successive elimination algorithm to solve this problem. We also derive a minimax lower bound to understand the fundamental limit on the sample complexity of this problem.
format Preprint
id arxiv_https___arxiv_org_abs_2203_16810
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Minimum mean-squared error estimation with bandit feedback
Ghosh, Ayon
Prashanth, L. A.
Sen, Dipayan
Gopalan, Aditya
Machine Learning
We consider the problem of sequentially learning to estimate, in the mean squared error (MSE) sense, a Gaussian $K$-vector of unknown covariance by observing only $m < K$ of its entries in each round. We propose two MSE estimators, and analyze their concentration properties. The first estimator is non-adaptive, as it is tied to a predetermined $m$-subset and lacks the flexibility to transition to alternative subsets. The second estimator, which is derived using a regression framework, is adaptive and exhibits better concentration bounds in comparison to the first estimator. We frame the MSE estimation problem with bandit feedback, where the objective is to find the MSE-optimal subset with high confidence. We propose a variant of the successive elimination algorithm to solve this problem. We also derive a minimax lower bound to understand the fundamental limit on the sample complexity of this problem.
title Minimum mean-squared error estimation with bandit feedback
topic Machine Learning
url https://arxiv.org/abs/2203.16810