Quantum agents in the Gym: a variational quantum algorithm for deep Q-learning

Abstract

Quantum machine learning (QML) has been identified as one of the key fields that could reap advantages from near-term quantum devices, next to optimization and quantum chemistry. Research in this area has focused primarily on variational quantum algorithms (VQAs), and several proposals to enhance supervised, unsupervised and reinforcement learning (RL) algorithms with VQAs have been put forward. Out of the three, RL is the least studied and it is still an open question whether VQAs can be competitive with state-of-the-art classical algorithms based on neural networks (NNs) even on simple benchmark tasks. In this work, we introduce a training method for parametrized quantum circuits (PQCs) that can be used to solve RL tasks for discrete and continuous state spaces based on the deep Q-learning algorithm. We investigate which architectural choices for quantum Q-learning agents are most important for successfully solving certain types of environments by performing ablation studies for a number of different data encoding and readout strategies. We provide insight into why the performance of a VQA-based Q-learning algorithm crucially depends on the observables of the quantum model and show how to choose suitable observables based on the learning task at hand. To compare our model against the classical DQN algorithm, we perform an extensive hyperparameter search of PQCs and NNs with varying numbers of parameters. We confirm that similar to results in classical literature, the architectural choices and hyperparameters contribute more to the agents' success in a RL setting than the number of parameters used in the model. Finally, we show when recent separation results between classical and quantum agents for policy gradient RL can be extended to inferring optimal Q-values in restricted families of environments.

Version accepted in Quantum journal, 16+3 pages, 14 figures

References in corpus (3)

• Quantum Wasserstein Generative Adversarial Networks
• Learning Unitaries by Gradient Descent
• Parametrized quantum policies for reinforcement learning

Cited by in corpus (39)

• Challenges and Opportunities in Quantum Machine Learning
• Quantum machine learning beyond kernel methods
• Quantum Computing for High-Energy Physics: State of the Art and Challenges. Summary of the QC4HEP Working Group
• Challenges and Opportunities in Quantum Optimization
• Variational Quantum Reinforcement Learning via Evolutionary Optimization
• Theoretical Guarantees for Permutation-Equivariant Quantum Neural Networks
• Shadows of quantum machine learning
• Quantum algorithms applied to satellite mission planning for Earth observation
• Hybrid quantum ResNet for car classification and its hyperparameter optimization
• Quantum Architecture Search via Deep Reinforcement Learning
• Quantum Deep Reinforcement Learning for Robot Navigation Tasks
• Quantum Reinforcement Learning via Policy Iteration
• Uncovering Instabilities in Variational-Quantum Deep Q-Networks
• QADQN: Quantum Attention Deep Q-Network for Financial Market Prediction
• Let Quantum Neural Networks Choose Their Own Frequencies
• Evolutionary Quantum Architecture Search for Parametrized Quantum Circuits
• Near-Optimal Quantum Algorithms for Multivariate Mean Estimation
• A hybrid classical-quantum approach to speed-up Q-learning
• Playing Atari with Hybrid Quantum-Classical Reinforcement Learning
• VQC-Based Reinforcement Learning with Data Re-uploading: Performance and Trainability
• Neural quantum kernels: training quantum kernels with quantum neural networks
• Hyperparameter Importance of Quantum Neural Networks Across Small Datasets
• Forecasting steam mass flow in power plants using the parallel hybrid network
• QTN-VQC: An End-to-End Learning framework for Quantum Neural Networks
• Guided-SPSA: Simultaneous Perturbation Stochastic Approximation assisted by the Parameter Shift Rule
• Hype or Heuristic? Quantum Reinforcement Learning for Join Order Optimisation
• Variational Quantum Circuit Design for Quantum Reinforcement Learning on Continuous Environments
• Quantum algorithms for multivariate Monte Carlo estimation
• Framework for Learning and Control in the Classical and Quantum Domains
• Digital Quantum Simulation and Circuit Learning for the Generation of Coherent States
• Model-based Offline Quantum Reinforcement Learning
• BCQQ: Batch-Constraint Quantum Q-Learning with Cyclic Data Re-uploading
• Quantum framework for Reinforcement Learning: Integrating Markov decision process, quantum arithmetic, and trajectory search
• Learning Fourier series with parametrized quantum circuits
• Transfer Learning Analysis of Variational Quantum Circuits
• Expressiveness of Commutative Quantum Circuits: A Probabilistic Approach
• Warm-Start Variational Quantum Policy Iteration
• QAISim: A Toolkit for Modeling and Simulation of AI in Quantum Cloud Computing Environments
• Trainability issues in quantum policy gradients