Personalized Automated Machine Learning

Cedric Kulbach, Patrick Philipp, Steffen Thoma
European Conference on Artificial Intelligence
Automated Machine Learning (AutoML) is the challenge of finding machine learning models with high predictive performance without the need for specialized data scientists. Existing approaches optimize a pipeline of pre-processing, feature engineering, model selection and hyperparameter optimization, and assume that the user is fully aware of the choice of the underlying metric (such as precision, recall or F1-measure). However, end-users are often unaware of the actual implications of choosing a metric, as the resulting models often significantly vary in their predictions. In this work, we propose a framework to personalise AutoML for individual end-users by learning a designated ranking model from pairwise user preferences and using the latter as the metric function for state-of-the-art AutoML systems. Given a set of possible metrics, we generate candidate models by repeatedly running AutoML with combinations of the former and have the user choose between pairs of resulting models. We use RankNet to learn a personalized ranking function for the end-user, which is used as loss function for final run of a standard AutoML system. To evaluate our proposed framework we define three preferences a user could pursue and show that a ranking model is able to learn these preferences from pairwise comparisons. Furthermore, by changing the metric function of AutoML we show that a personalized preference is able to improve machine learning pipelines. We evaluated the ability of learning a personalized preference and the entire framework on several OpenML multi-class classification datasets.
Download .bib
Download .bib
Eingetragen von
Cedric Kulbach