Quantum reinforcement learning

Abstract

The key approaches for machine learning, especially learning in unknown probabilistic environments are new representations and computation mechanisms. In this paper, a novel quantum reinforcement learning (QRL) method is proposed by combining quantum theory and reinforcement learning (RL). Inspired by the state superposition principle and quantum parallelism, a framework of value updating algorithm is introduced. The state (action) in traditional RL is identified as the eigen state (eigen action) in QRL. The state (action) set can be represented with a quantum superposition state and the eigen state (eigen action) can be obtained by randomly observing the simulated quantum state according to the collapse postulate of quantum measurement. The probability of the eigen action is determined by the probability amplitude, which is parallelly updated according to rewards. Some related characteristics of QRL such as convergence, optimality and balancing between exploration and exploitation are also analyzed, which shows that this approach makes a good tradeoff between exploration and exploitation using the probability amplitude and can speed up learning through the quantum parallelism. To evaluate the performance and practicability of QRL, several simulated experiments are given and the results demonstrate the effectiveness and superiority of QRL algorithm for some complex problems. The present work is also an effective exploration on the application of quantum computation to artificial intelligence.

13 pages, 7 figures, Latex

References in corpus (8)

• Quantum Mechanics helps in searching for a needle in a haystack
• Experimental realization of Shor's quantum factoring algorithm using nuclear magnetic resonance
• Tight bounds on quantum searching
• Implementation of a Quantum Search Algorithm on a Nuclear Magnetic Resonance Quantum Computer
• Grover's search algorithm: An optical approach
• Quantum Robots and Environments
• Quantum Robot: Structure, Algorithms and Applications
• Quantum mechanics helps in learning for more intelligent robot

Cited by in corpus (74)

• Noisy intermediate-scale quantum (NISQ) algorithms
• Quantum control theory and applications: A survey
• Quantum-enhanced machine learning
• Quantum computing for finance
• Experimental quantum speed-up in reinforcement learning agents
• Fidelity-based Probabilistic Q-learning for Control of Quantum Systems
• Variational Quantum Reinforcement Learning via Evolutionary Optimization
• Quantum Inference on Bayesian Networks
• Quantum Machine Learning: from physics to software engineering
• Basic protocols in quantum reinforcement learning with superconducting circuits
• Reconstruction of a Photonic Qubit State with Reinforcement Learning
• Robust Learning Control Design for Quantum Unitary Transformations
• Realization of a quantum autoencoder for lossless compression of quantum data
• Quantum machine learning and quantum biomimetics: A perspective
• Quantum Machine Learning for Finance
• Incoherent Control of Locally Controllable Quantum Systems
• Multiqubit and multilevel quantum reinforcement learning with quantum technologies
• Quantum Reinforcement Learning via Policy Iteration
• Uncovering Instabilities in Variational-Quantum Deep Q-Networks
• QADQN: Quantum Attention Deep Q-Network for Financial Market Prediction
• A quantum system control method based on enhanced reinforcement learning
• Reinforcement learning for semi-autonomous approximate quantum eigensolver
• Retrieving Quantum Information with Active Learning
• Entangled-photon decision maker
• Cyber-Physical Defense in the Quantum Era
• A duplication-free quantum neural network for universal approximation
• Enhanced Quantum Synchronization via Quantum Machine Learning
• Opening the Black Box Inside Grover's Algorithm
• Predominant Aspects on Security for Quantum Machine Learning: Literature Review
• Quantum reinforcement learning in continuous action space
• Measurement-based adaptation protocol with quantum reinforcement learning in a Rigetti quantum computer
• Quantum Bandits
• Quantum adaptive agents with efficient long-term memories
• Quantum spectral clustering algorithm for unsupervised learning
• Benefits of Open Quantum Systems for Quantum Machine Learning
• A hybrid classical-quantum approach to speed-up Q-learning
• Deep reinforcement learning for robust quantum optimization
• Playing Atari with Hybrid Quantum-Classical Reinforcement Learning
• Quantum machine learning with glow for episodic tasks and decision games
• Detecting Violations of Differential Privacy for Quantum Algorithms
• Quantum Algorithms for Reinforcement Learning with a Generative Model
• Quantum Search Approaches to Sampling-Based Motion Planning
• Optimization by a quantum reinforcement algorithm
• Machine Learning for Estimation and Control of Quantum Systems
• Quantum Computing for Software Engineering: Prospects
• Quantitative Predictions in Quantum Decision Theory
• SEQUENT: Towards Traceable Quantum Machine Learning using Sequential Quantum Enhanced Training
• Quantum reinforcement learning in the presence of thermal dissipation
• EP-PQM: Efficient Parametric Probabilistic Quantum Memory with Fewer Qubits and Gates
• Quantum autoencoders using mixed reference states
• Comparative Analysis of Quantum and Classical Support Vector Classifiers for Software Bug Prediction: An Exploratory Study
• A Quantum Implementation Model for Artificial Neural Networks
• A model of interacting quantum neurons with a dynamic synapse
• Quantum Computing for Data Centric Engineering and Science
• Quantum framework for Reinforcement Learning: Integrating Markov decision process, quantum arithmetic, and trajectory search
• BCQQ: Batch-Constraint Quantum Q-Learning with Cyclic Data Re-uploading
• Contextual Quantum Neural Networks for Stock Price Prediction
• Quantum Neural Networks: Concepts, Applications, and Challenges
• Dropout is all you need: robust two-qubit gate with reinforcement learning
• Adaptive Random Quantum Eigensolver
• Experimental semi-autonomous eigensolver using reinforcement learning
• Deep Reinforcement Learning with Quantum-inspired Experience Replay
• Unitary Evolutions Sourced By Interacting Quantum Memories: Closed Quantum Systems Directing Themselves Using Their State Histories
• Control of Overfitting with Physics
• The Holy Grail of Quantum Artificial Intelligence: Major Challenges in Accelerating the Machine Learning Pipeline
• Geometrical versus time-series representation of data in quantum control learning
• Intelligent Trajectory Planning in UAV-mounted Wireless Networks: A Quantum-Inspired Reinforcement Learning Perspective
• QZNs: Quantum Z-numbers
• Quantum spatial best-arm identification via quantum walks
• Multi-Armed Bandits and Quantum Channel Oracles
• Exploring fixed points and eigenstates of quantum systems with reinforcement learning
• Noise-Resilient Quantum Reinforcement Learning
• Verifiable cloud-based variational quantum algorithms
• Several recent developments in estimation and robust control of quantum systems