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

There is a consensus that focusing only on accuracy in searching for optimalmachine learning models amplifies biases contained in the data, leading tounfair predictions and decision supports. Recently, multi-objectivehyperparameter optimization has been proposed to search for machine learningmodels which offer equally Pareto-efficient trade-offs between accuracy andfairness. Although these approaches proved to be more versatile thanfairness-aware machine learning algorithms -- which optimize accuracyconstrained to some threshold on fairness -- they could drastically increasethe energy consumption in the case of large datasets. In this paper we proposeFanG-HPO, a Fair and Green Hyperparameter Optimization (HPO) approach based onboth multi-objective and multiple information source Bayesian optimization.FanG-HPO uses subsets of the large dataset (aka information sources) to obtaincheap approximations of both accuracy and fairness, and multi-objectiveBayesian Optimization to efficiently identify Pareto-efficient machine learningmodels. Experiments consider two benchmark (fairness) datasets and two machinelearning algorithms (XGBoost and Multi-Layer Perceptron), and provide anassessment of FanG-HPO against both fairness-aware machine learning algorithmsand hyperparameter optimization via a multi-objective single-sourceoptimization algorithm in BoTorch, a state-of-the-art platform for BayesianOptimization.

Fair and Green Hyperparameter Optimization via Multi-objective and Multiple Information Source Bayesian Optimization