Improving Automated Machine-Learning Systems through Green AI

Abstract

Automated machine learning (AutoML), which aims to facilitate the design and optimization of machine-learning models with reduced human effort and expertise, is a research field with significant potential to drive the development of artificial intelligence in science and industry. However, AutoML also poses challenges due to its resource and energy consumption and environmental impact, aspects that have often been overlooked. This paper predominantly centers on the sustainability implications arising from computational processes within the realm of AutoML. Within this study, a proof of concept has been conducted using the widely adopted Scikit-learn library. Energy efficiency metrics have been employed to fine-tune hyperparameters in both Bayesian and random search strategies, with the goal of enhancing the environmental footprint. These findings suggest that AutoML can be rendered more sustainable by thoughtfully considering the energy efficiency of computational processes. The obtained results from the experimentation are promising and align with the framework of Green AI, a paradigm aiming to enhance the ecological footprint of the entire AutoML process. The most suitable proposal for the studied problem, guided by the proposed metrics, has been identified, with potential generalizability to other analogous problems.

Sustainable Engineering of Machine Learning-Enabled Systems: A Systematic Mapping Study

“…Tornede et al [167] and Frey et al [159] argue for a holistic assessment in AutoML evaluations that includes both efficiency and environmental considerations, promoting energy-aware techniques in Neural Architecture Search (NAS) and Hyper-parameter optimization (HPO) to lower energy and computational demands. Simultaneously, Castellanos-Nieves and García-Forte [53] and Probst et al [105] examine the performance of Bayesian optimization versus random search in AutoML's hyperparameter tuning, finding Bayesian optimization marginally better for refining certain models and recommending the integrating bandit-based methods to achieve a more robust implementation [106]. Jean-Quartier et al [166] look into the sustainability impacts of explainable artificial intelligence (XAI) in the development of ML models, focusing on the energy consumption aspect.…”