Title: Hybrid Quantum Classical Autoencoder for End-to-End Communication Systems
Time: 2024年9月25日 14：00-15：00
Venue: beat365正版唯一九龍湖校區信息大樓834
Speaker: Prof. Gan Zheng
Abstract:
Autoencoder has emerged to be a new to design end-to-end communication systems. It replaces traditional channel coding, modulation and signal processing modules with neural networks and therefore does not rely on model information of a specific communications system. However, it requires a large number of parameters to optimize, and its complexity increases quickly with the size of the input messages. In this talk, I will introduce a hybrid quantum-classical autoencoder that incorporates quantum neural networks in the original autoencoder structure. Through extensive evaluations, we observe that the new hybrid quantum-classical autoencoder can reduce the number of parameters by almost half, while still achieving similar or better block error rate, and therefore provides a low-complexity and more scalable autoencoder structure.
Bio: Professor Gan Zheng a Fellow of IEEE and IET. His current research focuses on machine learning and quantum computing for wireless communications. He has published over 200 papers in international journals and conferences, which have received more than 13,000 citations. He received six international best paper awards. He currently serves as an Associate Editor for IEEE Wireless Communications Letters and IEEE Transactions on Communications.
