• Corpus ID: 240354766

Sustainable AI: Environmental Implications, Challenges and Opportunities

  title={Sustainable AI: Environmental Implications, Challenges and Opportunities},
  author={Carole-Jean Wu and Ramya Raghavendra and Udit Gupta and Bilge Acun and Newsha Ardalani and Kiwan Maeng and Gloria Chang and Fiona Aga Behram and James Huang and Charles Bai and Michael K. Gschwind and Anurag Gupta and Myle Ott and Anastasia Melnikov and Salvatore Candido and David Brooks and Geeta Chauhan and Benjamin Lee and Hsien-Hsin S. Lee and Bugra Akyildiz and Maximilian Balandat and Joe Spisak and Ravi Kumar Jain and Michael G. Rabbat and Kim M. Hazelwood},
This paper explores the environmental impact of the super-linear growth trends for AI from a holistic perspective, spanning Data, Algorithms, and System Hardware. We characterize the carbon footprint of AI computing by examining the model development cycle across industry-scale machine learning use cases and, at the same time, considering the life cycle of system hardware. Taking a step further, we capture the operational and manufacturing carbon footprint of AI computing and present an end-to… 

Unraveling the hidden environmental impacts of AI solutions for environment

The different types of AI impacts are reviewed, then the different methodologies used to assess those impacts are presented, and how to apply life cycle assessment to AI services is shown.

ACT: designing sustainable computer systems with an architectural carbon modeling tool

This work proposes ACT, an architectural carbon footprint modeling framework, to enable carbon characterization and sustainability-driven early design space exploration, and constructs use cases, based on the three tenets of sustainable design, to highlight future methods that enable strong performance and efficiency scaling in an environmentally sustainable manner.

Aligning artificial intelligence with climate change mitigation

A framework to assess how AI affects GHG emissions and proposes approaches to align the technology with climate change mitigation is presented.

Data-Centric Green AI: An Exploratory Empirical Study

Evidence is shown that, by exclusively conducting modifications on datasets, energy consumption can be drastically reduced, often at the cost of a negligible or even absent accuracy decline, which calls for a research agenda that focuses on data-centric techniques to improve AI energy efficiency.

Sustainable Computing - Without the Hot Air

This paper distills the fundamental trends that determine computing’s carbon footprint and their implications for achieving sustainable computing.

A Holistic Assessment of the Carbon Footprint of Noor, a Very Large Arabic Language Model

A holistic assessment of the footprint of an extreme-scale language model, Noor, aiming to develop the largest multi-task Arabic language models–with up to 13B parameters–leveraging zero-shot generalisation to enable a wide range of downstream tasks via natural language instructions is proposed.

Understanding Scaling Laws for Recommendation Models

Empirical scaling laws for DLRM style recommendation models, in particular Click-Through Rate (CTR), show that model quality scales with power law plus constant in model size, data size and amount of compute used for training.

Towards Green AI with tensor networks - Sustainability and innovation enabled by efficient algorithms

This paper presents a promising tool for sustainable and thus Green AI: tensor networks (TNs), an established tool from multilinear algebra that has the capability to improve efficiency without compromising accuracy, and argues that better algorithms should be evaluated in terms of both accuracy and ef-fiency.

Edge-Native Intelligence for 6G Communications Driven by Federated Learning: A Survey of Trends and Challenges

An overview of the state-of-the-art of FL applications in key wireless technologies that will serve as a foundation to establish a firm understanding of the topic and offer a road forward for future research directions.

Zeus: Understanding and Optimizing GPU Energy Consumption of DNN Training

An optimization framework to navigate the tradeoff between energy consumption and performance optimization is proposed, Zeus, which uses an online exploration-exploitation approach in conjunction with just-in-time energy profiling, averting the need for expensive of-time measurements.



Artificial intelligence for sustainability: Challenges, opportunities, and a research agenda

Chasing Carbon: The Elusive Environmental Footprint of Computing

This work quantifies the carbon output of computer systems to show that most emissions related to modern mobile and data-center equipment come from hardware manufacturing and infrastructure, and outlines future directions for minimizing the environmental impact of computing systems.

Quantifying the Carbon Emissions of Machine Learning

This work presents their Machine Learning Emissions Calculator, a tool for the community to better understand the environmental impact of training ML models and concrete actions that individual practitioners and organizations can take to mitigate their carbon emissions.

Energy Usage Reports: Environmental awareness as part of algorithmic accountability

This paper takes analyses usually applied at the industrial level and makes them accessible for individual computer science researchers with an easy-to-use Python package, and demonstrates the use of these reports as part of model-choice in a machine learning context.

Towards the Systematic Reporting of the Energy and Carbon Footprints of Machine Learning

A framework is introduced that makes accounting easier by providing a simple interface for tracking realtime energy consumption and carbon emissions, as well as generating standardized online appendices, and creates a leaderboard for energy efficient reinforcement learning algorithms to incentivize responsible research.

A first look into the carbon footprint of federated learning

A rigorous model to quantify the carbon footprint of FL is proposed, hence facilitating the investigation of the relationship between FL design and carbon emissions, and an early-stage FL optimization problem is formalized enabling the community to consider the importance of optimizing the rate of CO2 emissions jointly to the accuracy of neural networks.

DTCO including Sustainability: Power-Performance-Area-Cost-Environmental score (PPACE) Analysis for Logic Technologies

This work evaluates the energy and water consumption as well as the greenhouse gas emissions trends from processing logic nodes from iN28 to iN3 with different scaling scenarios and finds environmental scores can be derived together with the established Power, Performance, Area, Cost metrics.

On the Opportunities and Risks of Foundation Models

This report provides a thorough account of the opportunities and risks of foundation models, ranging from their capabilities, to their applications, and what they are even capable of due to their emergent properties.

Challenges: environmental design for pervasive computing systems

It is argued that pervasive computing offers not only tremendous opportunities and exciting research challenges but also possible negative environmental impacts, particularly in terms of physical waste and energy consumption, and that an important challenge for pervasive computing is to develop research in new architectures, design methodologies, metrics, algorithms and operating systems to minimize these impacts.

Tackling Climate Change with Machine Learning

From smart grids to disaster management, high impact problems where existing gaps can be filled by ML are identified, in collaboration with other fields, to join the global effort against climate change.