English

Power Consumption Variation over Activation Functions

Machine Learning 2020-06-15 v1 Neural and Evolutionary Computing Machine Learning

Abstract

The power that machine learning models consume when making predictions can be affected by a model's architecture. This paper presents various estimates of power consumption for a range of different activation functions, a core factor in neural network model architecture design. Substantial differences in hardware performance exist between activation functions. This difference informs how power consumption in machine learning models can be reduced.

Keywords

Cite

@article{arxiv.2006.07237,
  title  = {Power Consumption Variation over Activation Functions},
  author = {Leon Derczynski},
  journal= {arXiv preprint arXiv:2006.07237},
  year   = {2020}
}
R2 v1 2026-06-23T16:16:45.743Z