Hyperparameter optimisation of Deep Neural Networks for time-series forecasting

Type of thesis: Masterarbeit / location: Dresden / Status of thesis: Theses in progress

Simon Pickert currently works on his Master Thesis, developing a hyperparameter optimisation strategy for Deep Neural Networks (DNNs). The aim is to find a systematic approach to find the most appropriate set of hyperparameters of a DNN for the forecasting and gap-filling of energy load time-series data.

TU
Universität
Max
Leibnitz-Institut
Helmholtz
Hemholtz