The Comparison of Meta-heuristic and Reinforcement Learning Approach to Implement a Given Qubit Logic

We introduce the detailed comparison of meta-heuristic and reinforcement learning algorithms implementation in the area of quantum computations on the example of effective optimization of quantum control scheme to produce quantum logic gates with maximum fidelity to their theoretical counterpart. In particular, we compare the decision making process of the Genetic Algorithm (GA) as a meta-heuristic algorithm and Proximal Policy Optimization (PPO) as a reinforcement learning algorithm. We provide the comparison via the t-SNE and UMAP dimensionality reduction algorithms and analyze solution exploration process for each algorithm based on the reduced representation of generated solutions during training. Besides, a detailed review of the investigated problems and the algorithms used in the paper is given.