Full-Cycle Energy Consumption Benchmark for Low-Carbon Computer Vision

2021/08/30 by Bo Li, Li, Bo, Xinyang Jiang +15
Computer Science · #Advanced Image and Video Retrieval Techniques #Advanced Neural Network Applications #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Visual Attention and Saliency Detection #cs.CV #cs.LG

paper · pdf · doi:10.48550/arxiv.2108.13465

ArXiv Preprint

openalex publication_date 2021/08/30 · arxiv created 2021/10/12 · arxiv updated 2021/10/13 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

The energy consumption of deep learning models is increasing at a breathtaking rate, which raises concerns due to potential negative effects on carbon neutrality in the context of global warming and climate change. With the progress of efficient deep learning techniques, e.g., model compression, researchers can obtain efficient models with fewer parameters and smaller latency. However, most of the existing efficient deep learning methods do not explicitly consider energy consumption as a key performance indicator. Furthermore, existing methods mostly focus on the inference costs of the resulting efficient models, but neglect the notable energy consumption throughout the entire life cycle of the algorithm. In this paper, we present the first large-scale energy consumption benchmark for efficient computer vision models, where a new metric is proposed to explicitly evaluate the full-cycle energy consumption under different model usage intensity. The benchmark can provide insights for low carbon emission when selecting efficient deep learning algorithms in different model usage scenarios.

Citations

• FitNets: Hints for Thin Deep Nets
• Distilling the Knowledge in a Neural Network
• Network Trimming: A Data-Driven Neuron Pruning Approach towards Efficient Deep Architectures
• Neural Architecture Search with Reinforcement Learning
• Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfer
• Large-Scale Evolution of Image Classifiers
• Like What You Like: Knowledge Distill via Neuron Selectivity Transfer
• Paraphrasing Complex Network: Network Compression via Factor Transfer
• Post-training 4-bit quantization of convolution networks for rapid-deployment
• ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Once-for-All: Train One Network and Specialize it for Efficient Deployment
• pymoo: Multi-objective Optimization in Python
• Efficient Deep Learning: A Survey on Making Deep Learning Models Smaller, Faster, and Better
• Green AI

Related

• Double-Exponential Increases in Inference Energy: The Cost of the Race\n for Accuracy
• Accuracy is not the only Metric that matters: Estimating the Energy Consumption of Deep Learning Models
• Estimating Deep Learning energy consumption based on model architecture and training environment
• Watt For What: Rethinking Deep Learning's Energy-Performance Relationship
• Carbontracker: Tracking and Predicting the Carbon Footprint of Training Deep Learning Models