With the growing availability of large-scale datasets, and the popularization of affordable storage and computational capabilities, the energy consumed by AI is becoming a growing concern. To address this issue, in recent years, studies have focused on demonstrating how AI energy efficiency can be improved by tuning the model training strategy. Nevertheless, how modifications applied to datasets can impact the energy consumption of AI is still an open question.To fill this gap, in this exploratory study, we evaluate if datacentric approaches can be utilized to improve AI energy efficiency. To achieve our goal, we conduct an empirical experiment, executed by considering 6 different AI algorithms, a dataset comprising 5,574 data points, and two dataset modifications (number of data points and number of features).Our results show evidence that, by exclusively conducting modifications on datasets, energy consumption can be drastically reduced (up to 92.16%), often at the cost of a negligible or even absent accuracy decline. As additional introductory results, we demonstrate how, by exclusively changing the algorithm used, energy savings up to two orders of magnitude can be achieved.In conclusion, this exploratory investigation empirically demonstrates the importance of applying data-centric techniques to improve AI energy efficiency. Our results call for a research agenda that focuses on data-centric techniques, to further enable and democratize Green AI.
We present the EAM toolkit for life cycle modelling and impact analysis in environmental assessments. The open source toolkit was specifically designed to support maintainability and verification of models within integrated assessments, and has been used in research and industry. The tool offers features to support complex Life Cycle Assessment, including dynamic and scenario modelling, uncertainty and sensitivity analysis, a flexible domain specific modelling language and a visual editor. In this introduction we present the main features of the toolkit, summarise the high-level components and illustrate its use.CCS Concepts: • Human-centered computing → Empirical studies in HCI; HCI theory, concepts and models.
We present the DIMPACT /dimpaekt/ tool for environmental assessments of digital services. The tool enables digital media providers to calculate energy consumption and associated environmental impact across the entire product system, including datacentres, networks and user devices. It is based on accepted standard methodologies and applies state-of-the-art research. The DIMPACT tool is used by major media organisations for environmental reporting and to support development of strategies to reduce environmental impact. It has significantly advanced the knowledge of carbon emissions of video streaming. The tool is part of the wider DIMPACT project of companies that collaborate to exchange knowledge, engage suppliers and expand the scope of the tool. In this text we provide an overview of workings of tool and its methodological foundation.CCS Concepts: • Human-centered computing → Empirical studies in HCI; HCI theory, concepts and models.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.