2012/08/18 by Chris Thornton, Frank Hutter, Thornton, Chris +5
Computer Science · #D.2.10 #FOS: Computer and information sciences #I.2.2 #I.2.6 #Machine Learning (cs.LG) #cs.LG
paper · pdf · doi:10.48550/arxiv.1208.3719
9 pages, 3 figures
arxiv created 2013/03/06 · arxiv updated 2013/03/08
Many different machine learning algorithms exist; taking into account each algorithm's hyperparameters, there is a staggeringly large number of possible alternatives overall. We consider the problem of simultaneously selecting a learning algorithm and setting its hyperparameters, going beyond previous work that addresses these issues in isolation. We show that this problem can be addressed by a fully automated approach, leveraging recent innovations in Bayesian optimization. Specifically, we consider a wide range of feature selection techniques (combining 3 search and 8 evaluator methods) and all classification approaches implemented in WEKA, spanning 2 ensemble methods, 10 meta-methods, 27 base classifiers, and hyperparameter settings for each classifier. On each of 21 popular datasets from the UCI repository, the KDD Cup 09, variants of the MNIST dataset and CIFAR-10, we show classification performance often much better than using standard selection/hyperparameter optimization methods. We hope that our approach will help non-expert users to more effectively identify machine learning algorithms and hyperparameter settings appropriate to their applications, and hence to achieve improved performance.