Quantum Neural Networks for Speech and Natural Language Processing (QuantumNN) Tutorial

QuantumNN is a half-day (https://ijcai-21.org/tutorials/) interdisciplinary tutorial that will push the boundaries of quantum machine learning. We aim to bridge the fields of quantum circuit learning , deep neural networks, and providing hand-on examples on speech and language processing. | Virtual Room Link | IJCAI Registeration |

Tentative Schedule

Target Audience

Researchers, graduate students, and practitioners who are interested in Quantum Circuit Learning with Basic Knowledge of linear algebra and neural network training (e.g., gradient and back-propagation).

This tutorial aims to serve as a short lecture for researchers and students to access the emergent field of quantum neural networks from the viewpoint of the artificial intelligence community. The contents of this tutorial will provide sufficient backgrounds for participants to understand the motivation, research progress, opportunities, and ongoing challenges in quantum neural network-based speech and natural language processing. The outline of this tutorial is as follows:

• I. Introduction and Motivation of Quantum Machine Learning and Quantum Computer
• Principles of Quantum Mechanics and Tensor Network Foundation;
• Introduction to Quantum Computing;
• Quantum Machine Learning and Quantum Neural Networks;
• II. The Fundamentals of Quantum Neural Networks
• III. The Applications of Quantum Neural Networks for Speech and Language Processing
• IV. Conclusion and Open Questions

Anticipated target audience (introductory, intermediate, advanced) as well as expected number of attendees Background:linear algebra, basic understanding of class neural networks for speech recognition and natural language processing. All technical details will be provided with references and clear illustration and explanation.

We will provide Quantum Simulation Support and hands-on exercise through an open-source repository for the audience base on 5 qubits IBM Q devices.

Guest Speakers

Organizers

Introduction

The research of quantum machine learning is an emerging field that has flourished with the rapid development of quantum computing. In particular, quantum neural networks (QNNs), similar to classical neural networks, have already been applied in many large-scale machine learning tasks such as automatic speech recognition, speech enhancement, and natural language understanding. Despite the hardware limitation on noisy intermediate-scale quantum (NISQ) devices (5–50 qubits), the QNN based deep architectures, such as a randomized quantum convolutional neural network (QCNN) and variational quantum circuit (VQC), can be set up to attain competitive empirical results in experiments of speech and language processing. Moreover, more secured data privacy can be ensured by applying QNN based models.

Through IJCAI’s flagship and influence in AI research, we believe this tutorial can create the synergies and reinforce the momentum in advanced research and novel applications based on quantum computing and machine learning. This tutorial will provide an overview of the fundamentals of quantum mechanics, quantum machine learning and quantum neural networks. Then, we introduce the related applications in speech recognition and natural language understanding. In more detail, in the introduction part, we briefly introduce basic concepts of quantum computing, quantum mechanics and necessary multi-linear algebra associated with quantum technology. In the second section, we will discuss QNNs, especially based on variational quantum circuits (VQC) for QNNs. Finally, we provide several examples of employing VQC-QNN for speech recognition and natural language understanding.

Demo

Title: Demonstration of Quantum Circuits Learning for Spoken Commands Recognition

• Quantum Machine Learning for Automatic Spoken-Term Recognition: Google Colab
• Quantum Speech Video

---

Scientific Committee



Hsi-Sheng Goan

National Taiwan University

---

References

1. "Quantum machine learning in feature Hilbert spaces," M Schuld et al. , Physical review letters, 2019
2. Kosuke Mitarai, Makoto Negoro, Masahiro Kitagawa, and Keisuke Fujii. Quantum circuit learning. Physical Review A, 98(3):032309, 2018.
3. Maria Schuld, Alex Bocharov, Krysta Svore, and Nathan Wiebe. Circuit-centric quantum classifiers. arXiv preprint arXiv:1804.00633, 2018.
4. Maria Schuld, Ville Bergholm, Christian Gogolin, Josh Izaac, and Nathan Killoran. Evaluating analytic gradients on quantum hardware. Physical Review A, 99(3):032331, 2019.
5. Samuel Yen-Chi Chen, Chao-Han Huck Yang, Jun Qi, Pin-Yu Chen, XiaoliMa, and Hsi-Sheng Goan. Variational quantum circuits for deep reinforcement learning.IEEE Access, 8:141007–141024, 2020
6. "Decentralizing feature extraction with quantum convolutional neural network for automatic speech recognition."" Proc. IEEE Intl. Conf. on Acoustic, Speech, and Signal Processing (ICASSP), Yang, C.H.H. et al., 2021
7. Samuel Yen-Chi Chen, Tzu-Chieh Wei, Chao Zhang, Haiwang Yu, and Shinjae Yoo. Quantum convolutional neural networks for high energy physics data analysis. arXiv preprint arXiv:2012.12177, 2020.
8. Samuel Yen-Chi Chen, Tzu-Chieh Wei, Chao Zhang, Haiwang Yu, and Shinjae Yoo. Hybrid quantum-classical graph convolutional network.arXiv preprint arXiv:2101.06189, 2021
9. Samuel Yen-Chi Chen, Shinjae Yoo, and Yao-Lung L Fang. Quantum long short-term memory. arXiv preprint arXiv:2009.01783, 2020.
10. Samuel Yen-Chi Chen and Shinjae Yoo. Federated quantum machine learning. Entropy, 23(4):460, 2021.
11. Samuel Yen-Chi Chen, Chih-Min Huang, Chia-Wei Hsing, and Ying-Jer Kao. An end-to-end trainable hybrid classical-quantum classifier. arXiv preprint arXiv:2102.02416, 2021.

Website theme is modified and inspired from the VIGIL workshop Series. Florian S. et al.