Prof. Dr. Dr. Lars Schmidt-Thieme: "Hyperparameter Optimization"
29. Nov
Im Rahmen des Forschungskolloquiums Wirtschaftsinformatik und Data Science referiert Prof. Dr. Dr. Lars Schmidt-Thieme von der Universität Hildesheim über "Hyperparameter Optimization".
Datum und Ort: 19. November 2018 12:15 Uhr C12.010
Abstract:
Hyperparameter optimization is a tedious task that has to be dealt with when applying machine learning to real-world problems. Sequential model-based optimization (SMBO), based on so-called surrogate models, has been employed to allow for faster and more direct hyperparameter optimization than good old grid search. A surrogate model is a machine learning regression model which is trained on the meta level instances in order to predict the performance of a model on a specific data set given the hyperparameter settings and data set descriptors. Gaussian processes, for example, make good surrogate models as they provide probability distributions over labels. In this talk I will provide a gentle introduction to hyperparameter optimization and some of the recent approaches, esp. when hyperparameters are learnt across data sets.