Green AI
2019/07/22 by Roy Schwartz, Schwartz, Roy, Jesse Dodge +5 · 2 voices · 66 citations
Computer Science · Engineering · #Green IT and Sustainability #Mobile Crowdsensing and Crowdsourcing #cs.CL #cs.CV #cs.CY #cs.LG #stat.ME
paper · pdf · doi:10.48550/arxiv.1907.10597
openalex publication_date 2019/07/22 · openalex created_date 2022/10/06 · openalex updated_date 2026/07/28
Abstract
The computations required for deep learning research have been doubling every few months, resulting in an estimated 300,000x increase from 2012 to 2018 [2]. These computations have a surprisingly large carbon footprint [38]. Ironically, deep learning was inspired by the human brain, which is remarkably energy efficient. Moreover, the financial cost of the computations can make it difficult for academics, students, and researchers, in particular those from emerging economies, to engage in deep learning research. This position paper advocates a practical solution by making efficiency an evaluation criterion for research alongside accuracy and related measures. In addition, we propose reporting the financial cost or "price tag" of developing, training, and running models to provide baselines for the investigation of increasingly efficient methods. Our goal is to make AI both greener and more inclusive---enabling any inspired undergraduate with a laptop to write high-quality research papers. Green AI is an emerging focus at the Allen Institute for AI.
Cited by
- Attentions Under the Microscope: A Comparative Study of Resource Utilization for Variants of Self-Attention
- Sparser, Faster, Lighter Transformer Language Models
- Big AI is accelerating the metacrisis: What can we do?
- Life-Cycle Emissions of AI Hardware: A Cradle-To-Grave Approach and Generational Trends
- Keyword Matters: Unveiling the Energy Sensitivity of On-Device LLM Prompting
- Compressing LLMs with MoP: Mixture of Pruners
- Efficient Jailbreak Mitigation Using Semantic Linear Classification in a Multi-Staged Pipeline
- CienaLLM: Generative Climate-Impact Extraction from News Articles with Autoregressive LLMs
- Scaling Laws for Energy Efficiency of Local LLMs
- Reconsidering Conversational Norms in LLM Chatbots for Sustainable AI
- Fine-Grained Energy Prediction For Parallellized LLM Inference With PIE-P
- The 2025 Foundation Model Transparency Index
- SGEMAS: A Self-Growing Ephemeral Multi-Agent System for Unsupervised Online Anomaly Detection via Entropic Homeostasis
- Energy-Efficient Vision Transformer Inference for Edge-AI Deployment
- Through the telecom lens: Are all training samples important?
- CascadedViT: Cascaded Chunk-FeedForward and Cascaded Group Attention Vision Transformer
- ERMoE: Eigen-Reparameterized Mixture-of-Experts for Stable Routing and Interpretable Specialization
- Chain of Summaries: Summarization Through Iterative Questioning
- Intelligence per Watt: Measuring Intelligence Efficiency of Local AI
- Agentic AI Sustainability Assessment for Supply Chain Document Insights
- Green AI: A systematic review and meta-analysis of its definitions, lifecycle models, hardware and measurement attempts
- A Metamorphic Testing Perspective on Knowledge Distillation for Language Models of Code: Does the Student Deeply Mimic the Teacher?
- Quantifying the Climate Risk of Generative AI: Region-Aware Carbon Accounting with G-TRACE and the AI Sustainability Pyramid
- Who Evaluates AI's Social Impacts? Mapping Coverage and Gaps in First and Third Party Evaluations
- Sustainability of Machine Learning-Enabled Systems: The Machine Learning Practitioner's Perspective
- PowerScale: Energy-Efficient Geo-Distributed Model Training with Federated Datacenter Power
- Extractive versus Generative Language Models for Political Conflict Text Classification
- Toward Carbon-Neutral Human AI: Rethinking Data, Computation, and Learning Paradigms for Sustainable Intelligence
- Simplicial Embeddings Improve Sample Efficiency in Actor-Critic Agents
- Task-Aware Reduction for Scalable LLM-Database Systems
- The Environmental Impacts of Machine Learning Training Keep Rising Evidencing Rebound Effect
- Contrastive Self-Supervised Learning at the Edge: An Energy Perspective
- Beyond Static Knowledge Messengers: Towards Adaptive, Fair, and Scalable Federated Learning for Medical AI
- Person-Centric Annotations of LAION-400M: Auditing Bias and Its Transfer to Models
- Fair Minimum Labeling: Efficient Temporal Network Activations for Reachability and Equity
- Energy Efficiency in Cloud-Based Big Data Processing for Earth Observation: Gap Analysis and Future Directions
- Dissecting Transformers: A CLEAR Perspective towards Green AI
- Digital and Technology-Enabled Approaches in Dietary Assessment: Addressing Bias, Error, and Feasibility in Population- and Community-Based Research
- Green Prompt Engineering: Investigating the Energy Impact of Prompt Design in Software Engineering
- Ground-Truthing AI Energy Consumption: Validating CodeCarbon Against External Measurements
- Modeling Decisions in Blockchain Analytics: A Leakage-Aware Evaluation of Tree-Based vs. Sequential Models
- End-to-end probabilistic hierarchical forecasting of large hierarchies via probabilistic top-down
- Lift What You Can: Green Online Learning with Heterogeneous Ensembles
- Towards Open-Ended Discovery for Low-Resource NLP
- An Analysis of Optimizer Choice on Energy Efficiency and Performance in Neural Network Training
- Investigating ReLoRA: Effects on the Learning Dynamics of Small Language Models
- Performance is not All You Need: Sustainability Considerations for Algorithms
- Neuromorphic Intelligence
- Green Federated Learning via Carbon-Aware Client and Time Slot Scheduling
- Profiling LoRA/QLoRA Fine-Tuning Efficiency on Consumer GPUs: An RTX 4060 Case Study
- Greener Deep Reinforcement Learning: Analysis of Energy and Carbon Efficiency Across Atari Benchmarks
- Holographic Knowledge Manifolds: A Novel Pipeline for Continual Learning Without Catastrophic Forgetting in Large Language Models
- Gradient Estimation Methods of Approximate Multipliers for High-Accuracy Retraining of Deep Learning Models
- The Two Tales of AI: A Global assessment of the environmental impacts of artificial intelligence from a multidimensional policy perspective
- Evaluating the Evaluators: Are readability metrics good measures of readability?
- Bites of Tomorrow: Personalized Recommendations for a Healthier and Greener Plate
- Is GPT-OSS Good? A Comprehensive Evaluation of OpenAI's Latest Open Source Models
- Energy Consumption in Parallel Neural Network Training
- Position: Ideas Should be the Center of Machine Learning Research
- Embedding Is (Almost) All You Need: Retrieval-Augmented Inference for Generalizable Genomic Prediction Tasks
- Model Compression vs. Adversarial Robustness: An Empirical Study on Language Models for Code
- XAutoLM: Efficient Fine-Tuning of Language Models via Meta-Learning and AutoML
- Mitigating the Carbon Footprint of Chatbots as Consumers
- Sustainable AI Training via Hardware-Software Co-Design on NVIDIA, AMD, and Emerging GPU Architectures
- The Carbon Cost of Conversation, Sustainability in the Age of Language Models
- What Makes Code Generation Ethically Sourced?
Discussions
Related