Quantum Machine Learning

We explore the integration of variational quantum circuits (VQCs) and Condition Value at Risk (CVaR) to optimize the kinematics and network connectivity of autonomous vehicles (AVs) under stochastic conditions, such as wireless channels and traffic dynamics.

Our study employs VQC-based multi-objective reinforcement learning (MORL) to simultaneously manage:

• Cell-association and autonomous driving policies for AVs on a multi-lane highway.

• Base stations (BSs) operating across RF and THz spectrums.

To achieve these goals, we formulate the problem as a multi-objective Markov decision process (MOMDP) and transform it into quantum eigen-states and eigen-actions using quantum circuits. This approach aims to maximize handoff-aware data rates, enhance traffic flow, and ensure collision avoidance.

Related publications

• Zijiang Yan, Hao Zhou, Jianhua Pei, Aryan Kaushik, Hina Tabassum, Ping Wang. CVaR-Based Variational Quantum Optimization for User Association in Handoff-Aware Vehicular Networks. IEEE International Conference on Communications (ICC’25), 2025.

• Zijiang Yan, Ramsundar Tanikella, Hina Tabassum. Optimizing Vehicular Networks with Variational Quantum Circuits-based Reinforcement Learning. IEEE International Conference on Computer Communications (INFOCOM’24 Poster), 2024.