Statistical physics and Deep Learning

Datum

05.04.2018

Zeit

14:00 - 15:00

Sprecher

Mykola Maksymenko

Zugehörigkeit

Weizmann Institute of Science

Sprache

en

Hauptthema

Physik

Andere Themen

Physik

Beschreibung

We witness the rise of deep neural networks architectures making an enormous boost in applications as diverse as speech recognition and computer vision. However, in the practical sense, Deep Neural Networks usually are massively redundant, handcrafted and badly optimized systems, build with a large deal of intuition instead of rigorous arguments. Physics has long been linked to machine learning as both fields deal with exponentially large spaces and complex cost-function landscapes. In my talk, I will outline a connection between statistical physics and learning from examples and give some practical examples of how it applies to problems in modern deep learning.

Letztmalig verändert: 05.04.2018, 09:37:33