Policy Gradient Approach to Compilation of Variational Quantum Circuits
This provides a more efficient compilation method for variational quantum circuits, though it appears incremental relative to existing reinforcement learning applications in quantum computing.
The authors tackled the problem of compiling quantum unitary transformations by proposing a policy gradient reinforcement learning method that reformulates optimization as probability distribution parameter tuning rather than variational gate angle optimization. They showed numerically that this approach outperforms gradient-free methods for both noiseless and noisy circuits with comparable resources and eliminates the need for separate registers and long-range interactions required by Hilbert-Schmidt test methods.
We propose a method for finding approximate compilations of quantum unitary transformations, based on techniques from policy gradient reinforcement learning. The choice of a stochastic policy allows us to rephrase the optimization problem in terms of probability distributions, rather than variational gates. In this framework, the optimal configuration is found by optimizing over distribution parameters, rather than over free angles. We show numerically that this approach can be more competitive than gradient-free methods, for a comparable amount of resources, both for noiseless and noisy circuits. Another interesting feature of this approach to variational compilation is that it does not need a separate register and long-range interactions to estimate the end-point fidelity, which is an improvement over methods which rely on the Hilbert-Schmidt test. We expect these techniques to be relevant for training variational circuits in other contexts.