Automated machine learning automates the process of selecting the algorithms and hyperparameters that most accurately suit a specific problem. AutoML is introduced as a means to facilitate machine learning to non-machine learning experts, since usually, to achieve competitive solutions, machine learning experts with extensive domain knowledge are needed. In real-world contexts, selecting the best algorithm and algorithm setup is time-consuming and not always possible based on the available expertise.
Ekkono’s AutoML framework facilitates the selection of hyperparameters that are individually tailored with respect to each machine learning algorithm. This process is based on the data and takes into consideration choices of pipeline architectures, to build a solution uniquely tailored for the specific problem. Through a series of optimization techniques, AutoML selects hyperparameters and algorithms that are suited to address the specific problem.
