Taking gradients through experiments: LSTMs and memory proximal policy optimization for black-box quantum control
This work addresses quantum control optimization, a domain-specific challenge in physics, with incremental improvements to existing methods.
The authors tackled the problem of black-box quantum control by proposing a reinforcement learning method using LSTM networks and a novel memory proximal policy optimization (MPPO) algorithm, achieving state-of-the-art results on several quantum control tasks with discrete and continuous parameters.
In this work we introduce the application of black-box quantum control as an interesting rein- forcement learning problem to the machine learning community. We analyze the structure of the reinforcement learning problems arising in quantum physics and argue that agents parameterized by long short-term memory (LSTM) networks trained via stochastic policy gradients yield a general method to solving them. In this context we introduce a variant of the proximal policy optimization (PPO) algorithm called the memory proximal policy optimization (MPPO) which is based on this analysis. We then show how it can be applied to specific learning tasks and present results of nu- merical experiments showing that our method achieves state-of-the-art results for several learning tasks in quantum control with discrete and continouous control parameters.