Exploring Hybrid quantum-classical Neural Networks with PyTorch and Qiskit

[kifumi]

Abstract

Let's try and explore Hybrid quantum-classical Neural Networks with PyTorch and Qiskit: https://qiskit.org/textbook/ch-machine-learning/machine-learning-qiskit-pytorch.html

Examples:

• Classify the numbers you wrote.
• Classify the numbers other than 0 and 1.
• Add more quantum layers to improve the algorithm.

Description

There is various method to integrate quantum computing and Machine Learning. We can learn the relatively simple way using PyTorch and Qiskit on Qiskit Textbook. Its quantum portion is very limited but we might start from adding quantum layer to classify the numbers other than 0 and 1 to improve it.

For demonstration purposes, it would be a good way to advertise quantum machine learning if we can create an application that can classify the numbers written with the mouse in real time.

Members

• @githubhandle
• @githubhandle - Slack: @slackhandle email: [email protected]
• Qiskit Coach: @kifumi @attp

Deliverable

https://he-s3.s3.amazonaws.com/media/sprint/qiskit-hackathon-korea/team/964302/7dedd10qiskit_hackathon__12.pdf

GitHub repo

https://github.com/yh08037/quantum-neural-network

[yh08037]

@kifumi Thank you for accepting our team! Our Slack usernames : github ids are as follows.

Eunchan Lee : purang2 Dohun Kim : yh08037 Daeheon Yoon : Greathoney Kwon Yoon : vhapfks

[Jaehoon-zx]

Hi, I am interested in this topic and look forward to contribute to the project. Can I join the team?

[yunseo-kim]

I'm very interested in this topic! I want to contribute to this project. Can I join in?

[kifumi]

Hi, everyone. Thank you for your interest. There are good examples of code using different quantum tools like TensorFlow quantum and Pennylane:
https://www.tensorflow.org/quantum/tutorials/mnist https://pennylane.ai/qml/demos/tutorial_quanvolution.html How about rewriting these code to qiskit + pytorch?