Womanium Quantum+AI 2024 Projects
Please review the participation guidelines here before starting the project.
Do NOT delete/ edit the format of this read.me file.
Include all necessary information only as per the given format.
- Maximum team size = 2
- While individual participation is also welcome, we highly recommend team participation :)
- All nationalities, genders, and age groups are welcome to participate in the projects.
- All team participants must be enrolled in Womanium Quantum+AI 2024.
- Everyone is eligible to participate in this project and win Womanium grants.
- All successful project submissions earn the Womanium Project Certificate.
- Best participants win Womanium QSL fellowships with Fraunhofer ITWM. Please review the eligibility criteria for QSL fellowships in the project description below.
All information in this section will be considered for project submission and judging.
Ensure your repository is public and submitted by August 9, 2024, 23:59pm US ET.
Ensure your repository does not contain any personal or team tokens/access information to access backends. Ensure your repository does not contain any third-party intellectual property (logos, company names, copied literature, or code). Any resources used must be open source or appropriately referenced.
Team Member 1:
- Full Name: Naman Bansal
- Womanium Program Enrollment ID :WQ24-nNEou3pBkOohlYI
In this project, I tackled 5 tasks. The first task(link), was just getting acquainted with Pennylane as a platform. The second(link) introduced Variational classifiers, and this was used to fit a parity function, and classify Iris dataset Images. For these, tasks I modified the optimizers, and the circuits to achieve the required accuracy in fewer steps, and get the function to converge quickly.
For the 3rd task(link), I used the technique of Quanvolutional Neural Networks, to classify the MNIST data images. To further compare results, I made a custom ResNet-Type architecture on the top of quanvoluted images, to leverage the benfits of classical computing.
For the 4th task(link), I designed a single qubit circuit made up of single qubit rotation gates to fit the parameters, in order to simulate the sine function. The MSE found on a tes dataset was 0.00.
For the 5th task(link), I approached the problem with two techniques, Quantum Neural Networks and Quanvolutional Neural Networks, and evaluated their results based on their respective F1 scores. For the QNN, I reduced image dimensionality using PCA and implemented a hybrid model combining classical and quantum layers. For the second method, I applied a Quanvolutional layer on the images, and then passed them as input to a classical model. These results were compared to the passing just the images as it is to the same model, that is without quantum pre-processing.
Presentation Deck: https://1drv.ms/p/s!AheCgXOoKvGetUfJFSE23zrdtkND?e=TEKhhv
Recording : https://drive.google.com/file/d/1m6kKl6JqGzvQqRdF2fkdiEYfWnl9ULAJ/view?usp=sharing