I am Hassan and since Nov. 2008, a Post-Doc with
Prof. John Lloyd at CSL/RSISE,
Australian National University, working on the project
General Architecture for Artificially Intelligent Agents. I got my Doctorate from the
Dept. of Computer Science at the
University of Illinois at Urbana-Champaign in July/October 2008.
e-mail: hmahmud42 at gmail etc.
Learning Environment Models for Partially Observable Reinforcement Learning Problems, Transfer Learning , Bayesian Machine Learning, Universal Prediction and application of Algorithmic Information Theory to Machine Learning.
Behaving intelligently when Black Swan events dominate your reward received. In many (most?) problems of interest in practice, what determines the quality of your behaviour is how well you hedge against truly unforseen, unmodeled and highly damaging events. This includes everything from a robot getting knocked over to financial market crashes. The goal of this work is to devise a coherent reinforcement learning theory for dealing with such events. Starting points: robust control, robust reinforcement learning, strategies adopted by traders such as Nassim N. Taleb in navigating financial crashes.