Embark on a captivating journey through the mathematical universe behind deep neural networks in this Rothschild Lecture delivered by Professor Helmut Bölcskei from ETH Zürich. Explore the theoretical underpinnings of deep neural networks, delving into functional analysis, harmonic analysis, complex analysis, approximation theory, dynamical systems, Kolmogorov complexity, optimal transport, and fractal geometry. Gain insights into the mathematical foundations that have led to breakthrough results in practical machine learning tasks such as image classification, image captioning, control-policy-learning for the board game Go, and protein structure prediction. Discover how these mathematical concepts contribute to the remarkable successes of deep neural networks in various applications.
The Mathematical Universe Behind Deep Neural Networks - Rothschild Lecture
Isaac Newton Institute for Mathematical Sciences via YouTube
Overview
Syllabus
Date: 10 December 2021 – 16:00 to
Taught by
Isaac Newton Institute for Mathematical Sciences
Related Courses
-
Deep Learning - Recurrent Neural Networks with TensorFlow
-
Mathematics for Deep Neural Networks: Advantages of Additional Layers - Lecture 3
-
Mathematics for Deep Neural Networks: Statistical Theory for Deep ReLU Networks - Lecture 4
-
Empirical Challenges to Theories of Deep Learning
-
Mathematics for Deep Neural Networks: Theory for Shallow Networks - Lecture 2 of 5
-
High-Dimensional Classification with Deep Neural Networks: Decision Boundaries
