Dr. Dorota Glowacka has applied for the title of docent at the Department and she will give a public demonstration lecture on Friday 11.4. 10:15 at Exactum B222. The title of the lecture is "Introduction to Reinforcement Learning". Welcome!
Abstract:
Reinforcement learning is an area of machine learning inspired by behaviorist psychology, concerned with how software agents ought to take actions in an environment so as to maximize some notion of cumulative reward.
Reinforcement learning differs from standard supervised learning in that correct input/output pairs are never presented, nor sub-optimal actions explicitly corrected. Further, there is a focus on on-line performance, which involves finding a balance between exploration (of uncharted territory) and exploitation (of current knowledge). The exploration vs. exploitation trade-off in reinforcement learning has been most thoroughly studied through the multi-armed bandit problem. Thus, reinforcement learning is particularly well suited to problems which include a long-term versus short-term reward trade-off. It has been applied successfully to various problems, including robot control, elevator scheduling, telecommunications, backgammon and checkers.
In this lecture, we introduce the basic concepts of reinforcement learning with a particular emphasis on the applicational aspect of the field.
Bio:
Dorota Glowacka obtained a PhD in linguistics from University College London in 2005 as well as a PhD in machine learning from University College London in 2012. Her main research interests in the field of linguistics are phonological language acquisition and the interaction of prosody and morphology. In the field of machine learning, her main interests lie in reinforcement learning and the interplay between machine learning and human-computer interaction. Over the past decade, she taught a range of subjects both in linguistics (such as "Practical Phonetics" or "Phonological Theory") and in computer science (such as "Introduction to Programming" or "Reinforcement Learning").