About NMI Thematic Programmes Other Events Ph.d Programmes Request for Participation Visitor and Local Information Funding Agencies Contact
National Mathematics Initiative (NMI)

2014 - 2015

Recent Advances in Reinforcement learning


March 23 - 28, 2015

Explaining intelligent behavior in biological organisms has been one of holy grails of artificial intelligence (AI) research. Reinforcement learning (RL) started out as a model of learning in biological systems and today has grown to be one of the important paradigms of intelligent system design drawing ideas from varied fields such as neuroscience, psychology, control theory, operations research and AI. In turn reinforcement learning has had significant impact in many domains it is the most popular model of learning in computational neuroscience and is used to explain different phenomena observed in the brain; reinforcement learning methods have been used to build AI agents in domains which were traditionally regarded as very hard, such as the game of Go; RL has changed the traditional approach to adaptive optimal control theory by introducing newer ways of modeling system dynamics; and in robotics, RL is the primary learning paradigm used for training autonomous agents. From the early beginnings as a theory of behavioral psychology, over three decades RL has grown into a mathematically sophisticated field with rigorous underpinnings drawn from different disciplines. This workshop will introduce the participants to the basic concepts of reinforcement learning as well as more recent exciting results in the field from the leaders in the community.

Supported by National Mathematics Initiative
(Department of Science & Technology, Government of India)




Home | Contact | About IISc