Frugal Machine Learning for Energy-efficient, and Resource-aware Artificial Intelligence

TL;DR

Frugal Machine Learning (FML) focuses on creating resource-efficient ML models suitable for edge and IoT devices by minimizing computational, energy, and data requirements through various innovative strategies.

Contribution

This paper provides a comprehensive overview of recent advancements, taxonomy, applications, and open challenges in Frugal Machine Learning, emphasizing its role in resource-constrained environments.

Findings

FML techniques enable incremental updates without full retraining.

Model compression and energy-efficient hardware are key enablers.

FML is crucial for smart environments with strict resource limitations.

Abstract

Frugal Machine Learning (FML) refers to the practice of designing Machine Learning (ML) models that are efficient, cost-effective, and mindful of resource constraints. This field aims to achieve acceptable performance while minimizing the use of computational resources, time, energy, and data for both training and inference. FML strategies can be broadly categorized into input frugality, learning process frugality, and model frugality, each focusing on reducing resource consumption at different stages of the ML pipeline. This chapter explores recent advancements, applications, and open challenges in FML, emphasizing its importance for smart environments that incorporate edge computing and IoT devices, which often face strict limitations in bandwidth, energy, or latency. Technological enablers such as model compression, energy-efficient hardware, and data-efficient learning techniques are discussed, along with adaptive methods including parameter regularization, knowledge distillation, and dynamic architecture design that enable incremental model updates without full retraining. Furthermore, it provides a comprehensive taxonomy of frugal methods, discusses case studies across diverse domains, and identifies future research directions to drive innovation in this evolving field.