Astrid Merckling

Astrid Merckling

Machine Learning Expert

Biography

I completed my PhD at the Institute of Intelligent Systems and Robotics (ISIR) of Sorbonne University, advised by Pr. Stéphane Doncieux, Dr. Nicolas Perrin-Gilbert and Dr. Alex Coninx. My PhD focuses on central challenges of reinforcement learning and deep learning techniques. Specifically, it uses Machine Learning to give unsupervised exploration skills to robots, in order to solve new tasks more efficiently with Reinforcement Learning. To do this, we developed two new state-of-the-art algorithms – SRLfD and XSRL – for State Representation Learning with image observations. My long-term goal is to extend these discovery and adaptivity skills far beyond robotics to make all types of systems more autonomous and intelligent.

Download my French CV

Download my English CV

Interests
  • Computer Vision
  • Deep Learning
  • Unsupervised Learning
  • Reinforcement Learning
  • Exploration
Education
  • PhD in Machine Learning and Robotics

    ISIR – Sorbonne University

Contact