ALIFE-2021 Tutorial on:

Behavioral and Cognitive Robotics: An Adaptive Perspective

Stefano Nolfi


The tutorial will illustrate how to develop autonomous robots displaying behavioral and cognitive skills through evolutionary and reinforcement learning methods. It focuses on model-free methods that enable the robots to develop their skills autonomously, from scratch, on the basis of a fitness or reward function that rate how well they are doing.

The objectives are:

The tutorial will be based on the material included in Nolfi (2021) freely available from A guide for the utilization of the software tools reviewed in the tutorial is available from


Nolfi S. (2021). Behavioral and Cognitive Robotics: An Adaptive Perspective. Roma, Italy: Institute of Cognitive Sciences and Technologies, National Research Council (CNR-ISTC). ISBN 9791220082372.

Salimans T., Ho J., Chen X., Sidor S. & Sutskever I. (2017). Evolution strategies as a scalable alternative to reinforcement learning. arXiv:1703.03864v2,

Schulman J., Wolski F., Dhariwal P., Radford A. & Klimov O. (2017). Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347,


Stefano Nolfi

Stefano Nolfi ( is a research director at the Institute of Cognitive Sciences and Technologies of the Italian National Research Council where he coordinates the Laboratory of Autonomous Robots and Artificial Life. His main research interest is in study of how embodied and situated agents can develop behavioural and cognitive skills autonomously by adapting to their task/environment phylogenetically and ontogenetically. He conducted pioneering research in the field of Artificial Life and is one of the founders of the Evolutionary Robotics. Stefano authored and co-authored about 200 peer-review articles, a monograph on Evolutionary Robotics published in 2020 published by MIT Press, and book on Evolution of Communication and Language in Embodied Agents published in 2010 by Springer Verlag. He also coordinated and participated to several international projects.