- Speaker: Dr. Jun Qi, Fudan University
- Title： MAIB-Talk-005: Quantum Machine Learning: Theoretical Foundations and Applications on NISQ Devices
- Date：9:00pm US East time, 01/07/2023
- Date：10:00am Beijing time, 01/08/2023
- Zoom ID：933 1613 9423
- Zoom PWD：416262
- Key words: synthetic methodology based library; GIT/PIX; protein–protein interaction inhibitor
Title: Quantum Machine Learning: Theoretical Foundations and Applications on NISQ Devices
Abstract:
Quantum machine learning (QML) is a pioneering research topic that integrates quantum computing and machine learning. With the recent development of quantum computing, we have witnessed the advent of the NISQ era, which admits that up to a few hundred qubits are available for our QML applications, especially based on variable quantum circuits (VQC). This talk begins with a review of our pioneering research on VQC-based QML methods in reinforcement learning, speech recognition, and natural language processing. Then, we describe the theoretical basis of VQC and improve the representation and generalization capabilities of VQC by proposing an end-to-end TTN-VQC model. Furthermore, we further characterize hybrid quantum-classical neural networks in the context of meta-learning.
量子机器学习（QML）是一个开拓性的研究课题，它整合了量子计算和机器学习。随着最近量子计算的发展，我们见证了NISQ时代的到来，该时代承认有多达几百个量子比特可用于我们的QML应用，特别是基于变量量子电路（VQC）。本讲座首先回顾了我们在强化学习、语音识别和自然语言处理方面基于VQC的QML方法的开创性研究。然后，我们描述了VQC的理论基础，并通过提出一个端到端的TTN-VQC模型来提高VQC的表示和概括能力。此外，我们在元学习的背景下进一步描述了混合量子-经典神经网络的特点。
Bio:
Dr. Qi Jun is currently an assistant professor in the Department of Electronic Engineering, School of Information Science and Engineering, Fudan University. He will receive his Ph.D. degree in 2022 from Georgia Tech’s School of Electrical and Computer Engineering under the advisors of Prof. Chin-Hui Lee and Prof. Xiaoli Ma. Previously, he obtained two master’s degrees in electrical engineering from Seattle University in Washington and Tsinghua University in 2013 and 2017, respectively. In addition, he has conducted research internships at Microsoft Research in Redmond, WA, Tencent AI Lab in WA, and Deep Learning Technology Center at MERL in MA, USA. Dr. Qi is the first prize winner of the Xanadu AI Quantum Machine Learning Competition in 2019, and his ICASSP paper on quantum speech recognition was nominated as the best paper candidate in 2022. In addition, he gave two tutorials on Quantum Neural Networks for Speech and Language Processing at IJCAI21 and ICASSP22. Qi Jun was published as the first author in top quantum information journals such as NPJ Quantum Information, a sub-journal of Nature, and IEEE Trans. Signal Processing, IEEE Trans. Audio, Speech, and Langauge Processing, a top signal processing journal.
祁均博士现为复旦大学信息科学与工程学院电子工程系助理教授。他于2022年在佐治亚理工学院的电子和计算机工程学院获得博士学位，指导老师是Chin-Hui Lee教授和Xiaoli Ma教授。此前，他分别于2013年和2017年在华盛顿西雅图大学和清华大学获得两个电子工程硕士学位。此外，他曾在华盛顿州雷德蒙德的微软研究院、华盛顿州的腾讯人工智能实验室和美国马萨诸塞州的MERL的深度学习技术中心进行研究实习。祁博士是2019年Xanadu AI量子机器学习竞赛的一等奖获得者，他关于量子语音识别的ICASSP论文被提名为2022年的最佳论文候选人。此外，他在IJCAI21和ICASSP22的会场上演讲了两个关于量子神经网络用于语音和语言处理的教程。祁均以第一作者发表在量子信息顶刊如Nature子刊NPJ Quantum Information，信号处理顶刊IEEE Trans. Signal Processing, IEEE Trans. Audio, Speech, and Langauge Processing等。
-
Huang, H.Y., Broughton, M., Mohseni, M., Babbush, R., Boixo, S., Neven, H. and McClean, J.R., 2021. Power of data in quantum machine learning. Nature communications, 12(1), pp.1-9.
-
Jun Qi, et al. Theoretical Error Performance Analysis for Variational Quantum Circuit Based Functional Regression. Nature Publishing Group, npj Quantum Information, accepted in Dec. 2022.
https://sites.google.com/site/uwjunqi/home
https://scholar.google.com/citations?user=7oZpnlkAAAAJ&hl=en
https://sites.nd.edu/quantum/research-session/research-session-11prof-jun-qi/
