Search for contacts, projects,
courses and publications

RNNAIssance - RNNAIssance

People

 

Schmidhuber J.

(Responsible)

Rauber P.

(Collaborator)

Schlag I.

(Collaborator)

van Steenkiste S.

(Collaborator)

Zuo X.

(Collaborator)

Abstract

Real brains are still superior to our award-winning Recurrent Neural Networks (RNNs) in many ways: They learn a predictive model of their initially unknown environment, and somehow use it for abstract planning and reasoning. They continually build on earlier acquired skills, becoming more and more general problem solvers able to deal with many diverse and complex tasks. Guided by Algorithmic Information Theory, we will build a revolutionary RNN-based AI (RNNAI) that does the same.

Additional information

Acronym
RNNAIssance
Start date
01.11.2016
End date
29.02.2020
Duration
39 Months
Funding sources
SNSF
Status
Ended
Category
Swiss National Science Foundation / Project Funding / Division II - Mathematics, Natural and Engineering Sciences