Auto-WEKA: Automated Selection and Hyper-Parameter Optimization of Classification Algorithms

There exists a large variety of machine learning algorithms; as most of these can be configured via hyper-parameters, there is a staggeringly large number of possible alternatives overall. There has been a considerable amount of previous work on choosing among learning algorithms and, separately, on optimizing hyper-parameters (mostly when these are continuous and very few in number) in a given use context. However, we are aware of no work that addresses both problems together. Here, we demonstrate the feasibility of using a fully automated approach for choosing both a learning algorithm and its hyper-parameters, leveraging recent innovations in Bayesian optimization. Specifically, we apply this approach to the full range of classifiers implemented in WEKA, spanning 3 ensemble methods, 14 meta-methods, 30 base classifiers, and a wide range of hyper-parameter settings for each of these. On each of 10 popular data sets from the UCI repository, we show classification performance better than that of complete cross-validation over the default hyper-parameter settings of our 47 classification algorithms. We believe that our approach, which we dubbed Auto-WEKA, will enable typical users of machine learning algorithms to make better choices and thus to obtain better performance in a fully automated fashion.