XanaduAI / QHack2023

QHack 2023
69 stars 15 forks source link

Quantum-Image-Processing_Qualition #14

Open danielbultrini opened 1 year ago

danielbultrini commented 1 year ago

Project Name: Quantum Image Processing and Classification Team Name: Qualition

Which challenges would you like to submit your project for? NVIDIA Challenge (By dealing with QML and images, we hope that the project will be of interest to NVIDIA) Quantum Computing today! (QPIXL is an algorithm that was published in May 2022) Visualization Challenge (By performing various unitary transformations and entanglement between images we hope that it will give an intuitive 'view' into these operations) Hybrid Quantum-Classical Challenge (The project involves using both classical autoencoders and decoders as well as quantum ML to extend the capabilities of current machines)

Power-Up plan and project description: Power-Up plan: NVIDIA We would like to apply to the NVIDIA power-up, as we will be attempting to train a full quantum pipeline from encoding to classification. This will involve encoding images using either a chunked amplitude embedding or if time allows, a new pennylane or qiskit implementation of the QPIXL framework for FRQI. Then an autoencoder pipeline will feed into a quantum neural network classifier. In parallel to this, we would also like to see if it is possible to run various transformations on the encoded image and decode the state back into an image, a little bit like quantum photoshop. The NVIDIA power-up would greatly help in all of the training and simulation tasks.

Amazon In a similar vein to the above, having access to the computational resources afforded by AWS would increase our throughput with respect to the training of the various subsystems of our project. Since we will be attempting to classify cancer cells from healthy cells, we hope that is a potential use case with real-life implications.

Project Link:

https://github.com/Qualition/Quantum-Image-Processing/tree/1ee15a9a881f62bcb57bc538150192fe068fc449

We allow Xanadu Quantum Technologies to share our email addresses with the Power-Up Sponsors for the purpose of facilitating the delivery of the Power-Ups. YES

(If applying for AWS’s credits) We have an AWS account, the ID is: 334094085528

KetpuntoG commented 1 year ago

Hey! :) The link does not work, If you are applying to them you must modify it before the deadline 💪 On the other hand, we don't need your ID, we just need to know that you have it (just write YES)

danielbultrini commented 1 year ago

Hi KetpuntoG, the link seems to work - I just tried it on incognito mode - maybe there was some temporary glitch? But thank you for letting me know! As for the Yes for amazon, I would like to change it, but don't want to edit the issue and lose out on the potential other power ups. Than you for letting me know though, and got all your work in this Hackathon!