Dr. Daoyi Dong

Abstract In this talk, a novel quantum reinforcement learning (QRL) method is proposed by combining quantum theory and reinforcement learning (RL). Inspired by the state superposition principle and quantum parallelism, a framework of a value-updating algorithm is introduced. The state (action) in traditional RL is identified as the eigen state (eigen action) in QRL. The state (action) set can be represented with a quantum superposition state, and the eigen state (eigen action) can be obtained by randomly observing the simulated quantum state according to the collapse postulate of quantum measurement. The probability of the eigen action is determined by the probability amplitude, which is updated in parallel according to rewards. Some related characteristics of QRL such as convergence, optimality, and balancing between exploration and exploitation are also analyzed, which shows that this approach makes a good tradeoff between exploration and exploitation using the probability amplitude and can speedup learning through the quantum parallelism. To evaluate the performance and practicability of QRL, several simulated experiments are given, and the results demonstrate the effectiveness and superiority of the QRL algorithm for some complex problems.

Biography: Daoyi Dong got my Ph.D. from University of Science and Technology of China in 2006. Then I joined the Institute of Systems Science, Chinese Academy of Science as a postdoc fellow. May 2008, as an associate professor he joined Institute of Cyber-Systems and Control, Zhejiang University. Now he is working as an Australian Research Council APD Fellow at University of New South Wales, Australia and also working as a visiting scholar at Princeton University. His research interests include quantum control and intelligent information processing.