[ContinualAI Reading group]: "Class-Incremental Learning with Generative Classifiers"

This Friday 21-05-2021, 5.30pm CEST, for the ContinualAI Reading Group, Gido M. van de Ven will present the paper:

Title: “Class-Incremental Learning with Generative Classifiers

Abstract: Incrementally training deep neural networks to recognize new classes is a challenging problem. Most existing class-incremental learning methods store data or use generative replay, both of which have drawbacks, while ‘rehearsal-free’ alternatives such as parameter regularization or bias-correction methods do not consistently achieve high performance. Here, we put forward a new strategy for class-incremental learning: generative classification. Rather than directly learning the conditional distribution p(y|x), our proposal is to learn the joint distribution p(x,y), factorized as p(x|y)p(y), and to perform classification using Bayes’ rule. As a proof-of-principle, here we implement this strategy by training a variational autoencoder for each class to be learned and by using importance sampling to estimate the likelihoods p(x|y). This simple approach performs very well on a diverse set of continual learning benchmarks, outperforming generative replay and other existing baselines that do not store data.

The event will be moderated by: Vincenzo Lomonaco

:pushpin: Eventbrite event: Eventbrite Link
:pushpin: Miscrosoft Teams: click here to join
:pushpin: Youtube Recording: ContinualAI RG: “Class-Incremental Learning with Generative Classifiers”

Looking forward to seeing you all there!

Here are the slides of the above presentation: presentations/CLAIreadingGroup_May2021.pdf at main · GMvandeVen/presentations · GitHub

Thanks again for the invitation and for your attention. If anyone has any remaining question, or if you’d like to discuss anything related to this work, I’d be very happy to hear from you! :smiley:

1 Like