Modeling Time-Dependent Systems using Dynamic Quantum Bayesian Networks

TL;DR

This paper introduces a quantum-enhanced dynamic Bayesian network framework for modeling and controlling time-dependent systems under uncertainty, leveraging quantum algorithms for faster inference, and demonstrates its implementation on IBM quantum hardware.

Contribution

It presents the first integration of quantum amplitude amplification with dynamic Bayesian networks for real-time system modeling and control.

Findings

Quantum DQBN achieves quadratic speedup over classical methods.

Successful implementation of DQBN on IBM Q hardware.

Performance comparison shows advantages over classical approaches.

Abstract

Advances in data collection using inexpensive sensors have enabled monitoring the performance of dynamic systems, and to implement appropriate control actions to improve their performance. Moreover, engineering systems often operate under uncertain conditions; therefore, the real-time decision-making framework should not only consider real-time sensor data processing but also several uncertainty sources that may impact the performance of dynamic systems. In this paper, we investigate the modeling of such time-dependent system behavior using a dynamic quantum Bayesian network (DQBN), which is the quantum version of a classical dynamic Bayesian network (DBN). The DBN framework has been extensively used in various domains for its ability to model stochastic relationships between random variables across time. The use of the quantum amplitude amplification algorithm provides quadratic speedup for inference and prediction in Bayesian networks. In this paper, we combine the modeling capabilities of DBN with the computational advantage of quantum amplitude amplification for efficient modeling and control of time-dependent systems. We implement the proposed DQBN framework on IBM Q hardware, and compare its performance with classical DBN implementation and the IBM Qiskit simulator.