Generated data and plots for the paper "Exploring the Accuracy -- Energy Trade-off in Machine Learning"

Abstract

Machine learning accounts for considerable global electricity demand and resulting environmental impact, as training a large deep-learning model produces 284 000kgs of the greenhouse gas carbon dioxide. In recent years, search-based approaches have begun to explore improving software to consume less energy. Machine learning is a particularly strong candidate for this because it is possible to trade off functionality (accuracy) against energy consumption, whereas with many programs functionality is simply a pass-or-fail constraint. We use a grid search and NSGA-II optimisation run to explore hyperparameter configurations for a multilayer perceptron (from scikit-learn) on five classification data sets, considering trade-offs of classification accuracy against training or inference energy (using the PyRAPL) and run times. This includes the generated data of energy, time, accuracy metrics for each dataset, and the full set of corresponding plots. Details for each file are in the enclosed readme.