EC-NAS: Energy Consumption Aware Tabular Benchmarks for Neural Architecture Search
arXiv:2210.06015 · doi:10.1109/ICASSP48485.2024.10448303
Abstract
Energy consumption from the selection, training, and deployment of deep learning models has seen a significant uptick recently. This work aims to facilitate the design of energy-efficient deep learning models that require less computational resources and prioritize environmental sustainability by focusing on the energy consumption. Neural architecture search (NAS) benefits from tabular benchmarks, which evaluate NAS strategies cost-effectively through precomputed performance statistics. We advocate for including energy efficiency as an additional performance criterion in NAS. To this end, we introduce an enhanced tabular benchmark encompassing data on energy consumption for varied architectures. The benchmark, designated as EC-NAS, has been made available in an open-source format to advance research in energy-conscious NAS. EC-NAS incorporates a surrogate model to predict energy consumption, aiding in diminishing the energy expenditure of the dataset creation. Our findings emphasize the potential of EC-NAS by leveraging multi-objective optimization algorithms, revealing a balance between energy usage and accuracy. This suggests the feasibility of identifying energy-lean architectures with little or no compromise in performance.
Accepted to be presented at the International Conference on Acoustics, Speech and Signal Processing (ICASSP-2024). Source code at https://github.com/saintslab/EC-NAS-Bench
References in corpus (4)
- Compute Trends Across Three Eras of Machine Learning
- Carbontracker: Tracking and Predicting the Carbon Footprint of Training Deep Learning Models
- Carbon Footprint of Selecting and Training Deep Learning Models for Medical Image Analysis
- Bag of Baselines for Multi-objective Joint Neural Architecture Search and Hyperparameter Optimization