SAMOP 2023 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
QI: Fachverband Quanteninformation
QI 3: Quantum Machine Learning
QI 3.3: Talk
Monday, March 6, 2023, 11:45–12:00, B305
Parameterized quantum circuits for reinforcement learning of classical rare dynamics — Alissa Wilms1,2, •Laura Ohff2,3, Andrea Skolik4,5, David A. Reiss1, Sumeet Khatri1, and Jens Eisert1,6,7 — 1Dahlem Center for Complex Quantum Systems, Freie Universität Berlin, Berlin, Germany — 2Porsche Digital GmbH, Ludwigsburg, Germany — 3Otto-Friedrich Universität Bamberg, Bamberg, Germany — 4Leiden University, Leiden, The Netherlands — 5Volkswagen Data:Lab, Munich, Germany — 6Fraunhofer Heinrich Hertz Institute, Berlin, Germany — 7Helmholtz-Zentrum Berlin für Materialien und Energie, Berlin, Germany
In the study of non-equilibrium or industrial systems, rare events are crucial for understanding the systems' behavior. Since they are atypical, one requires specific methods for sampling and generating rare event statistics in an automated and statistically meaningful way. We propose two quantum reinforcement learning (QRL) approaches to study rare dynamics of time-dependent systems and investigate their benefits over classical approaches based on neural networks. We investigate how architectural choices influence the successful learning by QRL agents and demonstrate that a QRL agent is capable of learning the rare dynamics of a random walker with using just a single qubit. Furthermore, we are able to numerically demonstrate an improved environment exploration during learning and a better performance in coping with environment scaling by the quantum agents in comparison to their classical counterparts.