Budget Learning via Bracketing

TL;DR

This paper introduces a novel budget learning framework using bracketings to minimize cloud data transmissions in mobile/IoT devices without sacrificing accuracy, supported by theoretical analysis and empirical validation.

Contribution

It proposes a new bracket-based formulation for budget learning, with theoretical PAC and VC analyses, and demonstrates improved empirical performance over existing gating methods.

Findings

The bracket-based approach reduces cloud usage while maintaining accuracy.

Theoretical analysis establishes PAC learnability and VC bounds for brackets.

Empirical results show superior performance compared to prior gating methods.

Abstract

Conventional machine learning applications in the mobile/IoT setting transmit data to a cloud-server for predictions. Due to cost considerations (power, latency, monetary), it is desirable to minimise device-to-server transmissions. The budget learning (BL) problem poses the learner's goal as minimising use of the cloud while suffering no discernible loss in accuracy, under the constraint that the methods employed be edge-implementable. We propose a new formulation for the BL problem via the concept of bracketings. Concretely, we propose to sandwich the cloud's prediction, $g,$ via functions $h^-, h^+$ from a `simple' class so that $h^- \le g \le h^+$ nearly always. On an instance $x$, if $h^+(x)=h^-(x)$, we leverage local processing, and bypass the cloud. We explore theoretical aspects of this formulation, providing PAC-style learnability definitions; associating the notion of budget learnability to approximability via brackets; and giving VC-theoretic analyses of their properties. We empirically validate our theory on real-world datasets, demonstrating improved performance over prior gating based methods.