Sessions | Days | Selection | Search | Updates | Downloads | Help
FM: Fall Meeting
FM 42: Poster: Quantum Computation
FM 42.1: Poster
Tuesday, September 24, 2019, 16:30–18:30, Tents
A generalized search algorithm for quantum reinforcement learning — •Sabine Wölk, Arne Hamann, and Hans J. Briegel — Institut für Theoretische Physik, Universität Innsbruck, 6020 Innsbruck, Austria
Grover’s search is a well-known example of a quantum algorithm that provides a computational speedup with respect to the best classical counterpart. Among its several applications, Grover’s search has been used also in quantum machine learning and specifically to speed up algorithms for reinforcement learning. However, the search space in reinforcement learning tasks is not necessarily static, that is, the length of a single entry, as well as the number of entries and the target space are flexible. This represents a critical issue for the exploration stage during learning. Here, we propose a generalization of Grover’s search algorithm to monotonically increasing search spaces that is beneficial to tackle this issue in quantum reinforcement learning.