Research

This department studies how intelligent agents are able to learn about their environment by interacting with it, through curiosity-driven learning, using approaches such as deep learning, hierarchical reinforcement learning and meta-learning.

The department also develops technologies for program synthesis or induction and for assisting software developers, using deep learning and other machine learning methods. We are focusing on the automated development of web pages (HTML / CSS) or the synthesis of new types of autoencoders / compressors.

These two research topics are connected by research on curiosity-driven agents whose environment is a computer. By learning to control this environment, the agents learn how to automatically develop software.

Team

Publications