On Quantum Natural Policy Gradients
This work addresses the incremental impact of quantum FIM in reinforcement learning for researchers in quantum machine learning, showing it is not generally superior to classical methods.
This research tackled the problem of whether using the quantum Fisher Information Matrix (FIM) improves reinforcement learning with Parameterized Quantum Circuits (PQCs) beyond contextual bandits, finding that quantum FIM preconditioning incurs larger approximation errors and does not guarantee better performance than classical FIM in broader contexts like Markov Decision Processes.
This research delves into the role of the quantum Fisher Information Matrix (FIM) in enhancing the performance of Parameterized Quantum Circuit (PQC)-based reinforcement learning agents. While previous studies have highlighted the effectiveness of PQC-based policies preconditioned with the quantum FIM in contextual bandits, its impact in broader reinforcement learning contexts, such as Markov Decision Processes, is less clear. Through a detailed analysis of Löwner inequalities between quantum and classical FIMs, this study uncovers the nuanced distinctions and implications of using each type of FIM. Our results indicate that a PQC-based agent using the quantum FIM without additional insights typically incurs a larger approximation error and does not guarantee improved performance compared to the classical FIM. Empirical evaluations in classic control benchmarks suggest even though quantum FIM preconditioning outperforms standard gradient ascent, in general it is not superior to classical FIM preconditioning.