Quantum Multi-Agent Reinforcement Learning via Variational Quantum Circuit Design

Yun, Won Joon; Kwak, Yunseok; Kim, Jae Pyoung; Cho, Hyunhee; Jung, Soyi; Park, Jihong; Kim, Joongheon

doi:10.1109/icdcs54860.2022.00151

Cited by 26 publications

(18 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Notably, UAVs can provide wireless communication service without a fixed terrestrial infrastructure to create mobile access networks, providing broader wireless coverage and real-time service [13], [14]. UAVs can rapidly establish wireless connections with ground Parts of this research were appeared at IEEE International Conference on Distributed Computing Systems (ICDCS), Bologna, Italy, July 2022 [1].…”

Section: A Background and Motivationmentioning

confidence: 99%

“…Quantum MARL (QMARL) using quantum entanglement can handle the above non-stationary problem [20]. Building quantum neural networks (QNNs) using quantum computing (QC) enables efficient consumption of computing resources with fewer model parameters than conventional neural networks, resulting in more reliable and faster learning [1], [21], [22]. Existing studies have performed CTDE-based computation for the cooperation of multiple agents using a structure called CommNet [23].…”

Section: A Background and Motivationmentioning

confidence: 99%

“…We take care of this scalability issue, which is mentioned previously, in the quantum domain by using a structure in which a centralized critic network trains multiple actor networks iteratively [1]. This paper proposes a CTDE-based training algorithm, referred to as a QMACN algorithm utilizing the benefit of QC (i.e., quantum entanglement, efficient utilization of computing resource with fewer parameters).…”

Section: B Algorithm Design Rationalementioning

confidence: 99%

See 2 more Smart Citations

Quantum Multi-Agent Actor-Critic Networks for Cooperative Mobile Access in Multi-UAV Systems

Park¹,

Yun²,

Pyoung³

et al. 2023

Preprint

View full text Add to dashboard Cite

This paper proposes a novel quantum multi-agent actor-critic networks (QMACN) algorithm for autonomously constructing a robust mobile access system using multiple unmanned aerial vehicles (UAVs). For the cooperation of multiple UAVs for autonomous mobile access, multi-agent reinforcement learning (MARL) methods are considered. In addition, we also adopt the concept of quantum computing (QC) to improve the training and inference performances. By utilizing QC, scalability and physical issues can happen. However, our proposed QMACN algorithm builds quantum critic and multiple actor networks in order to handle such problems. Thus, our proposed QMACN algorithm verifies the advantage of quantum MARL with remarkable performance improvements in terms of training speed and wireless service quality in various data-intensive evaluations. Furthermore, we validate that a noise injection scheme can be used for handling environmental uncertainties in order to realize robust mobile access. Our data-intensive simulation results verify that our proposed QMACN algorithm outperforms the other existing algorithms.

show abstract

Section: A Background and Motivationmentioning

confidence: 99%

Section: A Background and Motivationmentioning

confidence: 99%

Section: B Algorithm Design Rationalementioning

confidence: 99%

See 1 more Smart Citation

Quantum Multi-Agent Actor-Critic Networks for Cooperative Mobile Access in Multi-UAV Systems

Park¹,

Yun²,

Pyoung³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Recently, the importance of software development tools for quantum machine learning (QML) is increasing. Nowadays, the QML software is one of the active research topics in modern noisy intermediate-scale quantum (NISQ) era [3]- [7]. Along with this fast QML research innovation, the relevant software development tools are widely used, such as Pen-nyLane, Qiskit, and torch-quantum.…”

Section: Introductionmentioning

confidence: 99%

EQuaTE: Efficient Quantum Train Engine for Dynamic Analysis via HCI-based Visual Feedback

Park¹,

Yun²,

Park³

et al. 2023

Preprint

View full text Add to dashboard Cite

This paper proposes an efficient quantum train engine (EQuaTE), a novel tool for quantum machine learning software which plots gradient variances to check whether our quantum neural network (QNN) falls into local minima (called barren plateaus in QNN). This can be realized via dynamic analysis due to undetermined probabilistic qubit states. Furthermore, our EQuaTE is capable for HCI-based visual feedback because software engineers can recognize barren plateaus via visualization; and also modify QNN based on this information.

show abstract

“…In this scenario, the reward is distributed to all agents concurrently by concatenating their state-actions pairs. A naïve implementation of a VQC version of CTDE is possible, as shown in [1]. However, such implementation causes the qubits to increase with the number of agents because when QRL is carried out via VQC, the stateaction pairs are represented by qubits.…”

mentioning

confidence: 99%

Quantum Multiagent Actor–Critic Neural Networks for Internet-Connected Multirobot Coordination in Smart Factory Management

Yun

Kim

Jung

et al. 2023

IEEE Internet Things J.

View full text Add to dashboard Cite

As one of the latest fields of interest in both academia and industry, quantum computing has garnered significant attention. Among various topics in quantum computing, variational quantum circuits (VQC) have been noticed for their ability to carry out quantum deep reinforcement learning (QRL). This paper verifies the potential of QRL, which will be further realized by implementing quantum multi-agent reinforcement learning (QMARL) from QRL, especially for Internet-connected autonomous multi-robot control and coordination in smart factory applications. However, the extension is not straightforward due to the non-stationarity of classical MARL. To cope with this, the centralized training and decentralized execution (CTDE) QMARL framework is proposed under the Internet connection. A smart factory environment with the Internet of Things (IoT)-based multiple agents is used to show the efficacy of the proposed algorithm. The simulation corroborates that the proposed QMARL-based autonomous multi-robot control and coordination performs better than the other frameworks.

show abstract

Quantum Multi-Agent Reinforcement Learning via Variational Quantum Circuit Design

Cited by 26 publications

References 11 publications

Quantum Multi-Agent Actor-Critic Networks for Cooperative Mobile Access in Multi-UAV Systems

Quantum Multi-Agent Actor-Critic Networks for Cooperative Mobile Access in Multi-UAV Systems

EQuaTE: Efficient Quantum Train Engine for Dynamic Analysis via HCI-based Visual Feedback

Quantum Multiagent Actor–Critic Neural Networks for Internet-Connected Multirobot Coordination in Smart Factory Management

Contact Info

Product

Resources

About