How to Allocate Resources For Features Acquisition?

TL;DR

This paper presents a method to optimally allocate resources for feature acquisition in noisy classification tasks, improving performance by non-uniform resource distribution and providing theoretical bounds and simulation validation.

Contribution

It introduces a novel approach for optimal resource allocation in noisy feature acquisition, with theoretical analysis and practical simulation results.

Findings

Non-uniform resource allocation can significantly improve classification accuracy.

Theoretical bounds quantify the benefits of optimized resource distribution.

Simulations confirm the effectiveness of the proposed method.

Abstract

We study classification problems where features are corrupted by noise and where the magnitude of the noise in each feature is influenced by the resources allocated to its acquisition. This is the case, for example, when multiple sensors share a common resource (power, bandwidth, attention, etc.). We develop a method for computing the optimal resource allocation for a variety of scenarios and derive theoretical bounds concerning the benefit that may arise by non-uniform allocation. We further demonstrate the effectiveness of the developed method in simulations.