Vijaykumar Gullapalli

Learn More
Acknowledgements Andrew Barto has been a constant source of inspiration to me throughout my tenure as a graduate student. His excellent technical expertise in an incredible range of disciplines makes him unique among researchers and unique as an advisor. I have beneted tremendously from Andy's guidance, enthusiasm, constructive criticism, and(More)
Dedicated to Mom and Dad, who gave me everything ACKNOWLEDGMENTS First I would like to thank my advisor, Andy Barto, for his continued guidance, inspiration and support, for sharing many of his insights, and for his constant eeort to make me a better writer. I owe much to him for the freedom he gave me to pursue my own interests. Thanks to Rich Sutton for(More)
The \forward modeling" approach of Jor-dan and Rumelhart has been shown to be applicable when supervised learning methods are to be used for solving reinforcement learning tasks. Because such tasks are natural candidates for the application of reinforcement learning methods, there is a need to evaluate the relative merits of these two learning methods on(More)
Reinforcement Learning methods based on approximating dynamic programming (DP) are receiving increased attention due to their utility in forming reactive control policies for systems embedded in dynamic environments. Environments are usually modeled as controlled Markov processes, but when the environment model is not known a priori, adaptive methods are(More)
We present a new algorithm for solving Markov decision problems that extends the modiied policy iteration algorithm of Puterman and Shin 6] in two important ways: 1) The new algorithm is asynchronous in that it allows the values of states to be updated in arbitrary order, and it does not need to consider all actions in each state while updating the policy.(More)
In this paper, a peg-in-hole insertion task is used as an example to illustrate the utility of direct as-sociative reinforcement learning methods for learning control under real-world conditions of uncertainty and noise. An associative reinforcement learning system has to learn appropriate actions in various situations through search guided by evaluative(More)