Ensemble Squared: A Meta AutoML System
This work addresses the problem of making machine learning more accessible to non-experts by improving the performance of automated machine learning systems, which benefits users without ML expertise.
This paper introduces Ensemble Squared (Ensemble^2), an AutoML system that combines the predictions of multiple existing state-of-the-art open-source AutoML systems. By leveraging the diversity in their search spaces and heuristics, Ensemble^2 achieves new state-of-the-art results on the OpenML tabular classification benchmark.
There are currently many barriers that prevent non-experts from exploiting machine learning solutions ranging from the lack of intuition on statistical learning techniques to the trickiness of hyperparameter tuning. Such barriers have led to an explosion of interest in automated machine learning (AutoML), whereby an off-the-shelf system can take care of many of the steps for end-users without the need for expertise in machine learning. This paper presents Ensemble Squared (Ensemble$^2$), an AutoML system that ensembles the results of state-of-the-art open-source AutoML systems. Ensemble$^2$ exploits the diversity of existing AutoML systems by leveraging the differences in their model search space and heuristics. Empirically, we show that diversity of each AutoML system is sufficient to justify ensembling at the AutoML system level. In demonstrating this, we also establish new state-of-the-art AutoML results on the OpenML tabular classification benchmark.