Automated machine learning (AutoML) aims to automate the machine learning process including feature engineering, model selection, and algorithm selection. The AutoML framework includes an optimizer to determine the search space and techniques for evaluation, an evaluator to measure the performance of learning tools, and a learning process. Key techniques for the optimizer include simple search spaces for evaluation and more complex derivative-free optimization. Techniques for the evaluator include direct evaluation on a validation set as well as early stopping and sub-sampling for faster evaluation.