Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Quantum Reinforcement Learning: Recent Advances and Future Directions

Jawaher Kaldari, Shehbaz Tariq, Saif Al-Kuwari, Samuel Yen-Chi Chen, Symeon Chatzinotas, Hyundong Shin

TL;DR

This survey reviews the field of quantum reinforcement learning (QRL), detailing its problem setting, taxonomy, and how variational quantum circuits enable near-term approaches. It introduces four architectures (QMARL, FERL, QVARL, QHRL) and three main algorithm classes (policy gradient, Q-learning with variational circuits, and actor–critic), with tutorials and benchmarking discussions. The paper also analyzes software frameworks (Qiskit, PennyLane, TensorFlow Quantum) and benchmarks, addressing hardware noise and the need for standardized datasets. It highlights applications across quantum control, error correction, sensing, architecture search, and autonomous systems, and outlines future directions toward hybrid quantum-classical computing and quantum-centric supercomputing.

Abstract

As quantum machine learning continues to evolve, reinforcement learning stands out as a particularly promising yet underexplored frontier. In this survey, we investigate the recent advances in QRL to assess its potential in various applications. While QRL has generally received less attention than other quantum machine learning approaches, recent research reveals its distinct advantages and transversal applicability in both quantum and classical domains. We present a comprehensive analysis of the QRL framework, including its algorithms, architectures, and supporting SDK, as well as its applications in diverse fields. Additionally, we discuss the challenges and opportunities that QRL can unfold, highlighting promising use cases that may drive innovation in quantum-inspired reinforcement learning and catalyze its adoption in various interdisciplinary contexts.

Quantum Reinforcement Learning: Recent Advances and Future Directions

TL;DR

This survey reviews the field of quantum reinforcement learning (QRL), detailing its problem setting, taxonomy, and how variational quantum circuits enable near-term approaches. It introduces four architectures (QMARL, FERL, QVARL, QHRL) and three main algorithm classes (policy gradient, Q-learning with variational circuits, and actor–critic), with tutorials and benchmarking discussions. The paper also analyzes software frameworks (Qiskit, PennyLane, TensorFlow Quantum) and benchmarks, addressing hardware noise and the need for standardized datasets. It highlights applications across quantum control, error correction, sensing, architecture search, and autonomous systems, and outlines future directions toward hybrid quantum-classical computing and quantum-centric supercomputing.

Abstract

As quantum machine learning continues to evolve, reinforcement learning stands out as a particularly promising yet underexplored frontier. In this survey, we investigate the recent advances in QRL to assess its potential in various applications. While QRL has generally received less attention than other quantum machine learning approaches, recent research reveals its distinct advantages and transversal applicability in both quantum and classical domains. We present a comprehensive analysis of the QRL framework, including its algorithms, architectures, and supporting SDK, as well as its applications in diverse fields. Additionally, we discuss the challenges and opportunities that QRL can unfold, highlighting promising use cases that may drive innovation in quantum-inspired reinforcement learning and catalyze its adoption in various interdisciplinary contexts.
Paper Structure (41 sections, 39 equations, 4 figures, 5 tables)

This paper contains 41 sections, 39 equations, 4 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Preliminaries
  3. Reinforcement Learning
  4. Variational Quantum Circuit
  5. Quantum Reinforcement Learning
  6. Taxonomy
  7. Definition
  8. Software Frameworks
  9. QRL Architectures
  10. Quantum multi-agent reinforcement learning
  11. Free energy-based reinforcement learning
  12. Quantum variational autoencoder for reinforcement learning
  13. Quantum Hierarchical Reinforcement Learning
  14. QRL Algorithms
  15. Quantum policy gradient
  16. ...and 26 more sections

Figures (4)

  • Figure 1: Reinforcement Learning cycle where the agents recursively interact with their environment and learn by associating rewards with their actions.
  • Figure 2: General framework of a VQC, illustrating parameterized unitary operations for applications in optimization, learning, and quantum-enhanced tasks.
  • Figure 3: Taxonomy of Quantum Reinforcement Learning approaches
  • Figure 4: Illustration of QRL's transversal applicability, showcasing its potential to enhance learning and decision-making in both quantum-specific and classical application domains.