Engineering

Ecology

Deep neural networks

Biology

Variety (cybernetics)

Operating system

Computer science

Cloud computing

Paleontology

Greenhouse gas

Sociology

Data science

Artificial intelligence

Deep learning

Consumption (sociology)

Artificial neural network

Electricity

Machine learning

Footprint

Carbon footprint

Electrical engineering

Social science

Recent progress in hardware and methodology for training neural networks has ushered in a new generation of large networks trained on abundant data. These models have obtained notable gains in accuracy across many NLP tasks. However, these accuracy improvements depend on the availability of exceptionally large computational resources that necessitate similarly substantial energy consumption. As a result these models are costly to train and develop, both financially, due to the cost of hardware and electricity or cloud compute time, and environmentally, due to the carbon footprint required to fuel modern tensor processing hardware. In this paper we bring this issue to the attention of NLP researchers by quantifying the approximate financial and environmental costs of training a variety of recently successful neural network models for NLP. Based on these findings, we propose actionable recommendations to reduce costs and improve equity in NLP research and practice.

Energy and Policy Considerations for Deep Learning in NLP