Workshop
Abstract
In this presentation, we will introduce core concepts of modern machine learning, with a focus on different deep learning architectures (such as Convolutional Neural Networks, recurrent networks, and Transformers), as well as aspects of the machine learning life cycle, such as training, application, and evaluation of deep learning models – including approaches to quantify and tackle the challenge of generalization beyond the training data distribution.
Topics
Overview about the necessary Python tools (Jupyter Notebook, PyTorch, Pandas, …)
Loading and exploring a data set
Defining and training a deep neural network with PyTorch
Evaluating and interpreting model performance
About the Speaker
Felix Putze
Researcher at University of Bremen
Biography
Felix Putze graduated in 2014 with a PhD at the Karlsruhe Institute of Technology on the topic “Adaptive Cognitive Interaction Systems”. Currently, he is working as a group leader at the Cognitive Systems Lab. His research foci are on the use of multimodal, heterogeneous data for capturing and modeling cognitive states of people. One important use case is the improvement of human-machine-interaction through the development of adaptive interfaces, but he is also interested in basic research in cognitive science. He is Principal Investigator in the Collaborative Research Center EASE on the topic “Enabling Next-Level LabLinking Framework for Virtual Experiment Spaces to Study Everyday Activities” as well as the research unit Lifespan AI with the topics “Lifespan Knowledge Representation” and “Hybrid Universal Embedding”.