English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Modern AI practices all strive towards the same goal: better results. In thecontext of deep learning, the term "results" often refers to the achievedaccuracy on a competitive problem set. In this paper, we adopt an idea from theemerging field of Green AI to consider energy consumption as a metric of equalimportance to accuracy and to reduce any irrelevant tasks or energy usage. Weexamine the training stage of the deep learning pipeline from a sustainabilityperspective, through the study of hyperparameter tuning strategies and themodel complexity, two factors vastly impacting the overall pipeline's energyconsumption. First, we investigate the effectiveness of grid search, randomsearch and Bayesian optimisation during hyperparameter tuning, and we find thatBayesian optimisation significantly dominates the other strategies.Furthermore, we analyse the architecture of convolutional neural networks withthe energy consumption of three prominent layer types: convolutional, linearand ReLU layers. The results show that convolutional layers are the mostcomputationally expensive by a strong margin. Additionally, we observediminishing returns in accuracy for more energy-hungry models. The overallenergy consumption of training can be halved by reducing the networkcomplexity. In conclusion, we highlight innovative and promisingenergy-efficient practices for training deep learning models. To expand theapplication of Green AI, we advocate for a shift in the design of deep learningmodels, by considering the trade-off between energy efficiency and accuracy.

Uncovering Energy-Efficient Practices in Deep Learning Training: Preliminary Steps Towards Green AI