The ability of continual learning systems to transfer knowledge from
pre...
Continual Learning aims to bring machine learning into a more realistic
...
We introduce a new training paradigm that enforces interval constraints ...
One of the main arguments behind studying disentangled representations i...
In this paper we present the first safe system for full control of
self-...
In this work we are the first to present an offline policy gradient meth...
Modern generative models achieve excellent quality in a variety of tasks...
The problem of reducing processing time of large deep learning models is...
Continual learning (CL) – the ability to continuously learn, building on...
We develop a fast end-to-end method for training lightweight neural netw...
We introduce bio-inspired artificial neural networks consisting of neuro...
We propose a semi-supervised generative model, SeGMA, which learns a joi...
Hypernetworks mechanism allows to generate and train neural networks (ta...