Dynamics of on-line learning in two-layer neural networks in the presence of concept drift

Abstract

In this talk I will discuss our recent work on learning in non-stationary situations. A modelling scenario of neural networks learning under a random drift process is presented and results will be discussed. Particularly we will focus on the effects of weight decay as a forgetting mechanism.

Date
Aug 12, 2020 7:00 PM
Location
École de Physique des Houches
149 Chemin de la Côté, Les Houches, Rhône-Alpes 74310

A recording of the talk can be found below.

Avatar
Michiel Straat
Postdoctoral Research Group Leader “Lifelong Machine Learning for Physical Systems”

My research interests include Machine Learning for Physical Systems and the theory of Neural Networks.