ML for Dependable Systems: Beyond Black Boxes

Description

Deep Learning has revolutionized many fields including healthcare, smart farming, and criminal justice, but its "black box" nature poses significant challenges. In this lecture, we will explore how understanding and interpreting these complex models is crucial for their dependable application. We'll use as an example convolutional neural networks, the basic building blocks of deep learning, and delve into techniques for visualizing and interpreting neuron activity. By examining how neurons form circuits and tackle specific tasks, we will highlight why interpreting these networks is so challenging. We will cover the intricacies of neural networks, the importance of transparency, and the future of making AI decisions more understandable and trustworthy.

Period	8 Jul 2024
Event title	DENISE Summer School "Machine Learning and Optimization for Dependable Systems"
Event type	Other
Location	AustriaShow on map
Degree of Recognition	International