XanaduAI / QHack2022

QHack—The one-of-a-kind quantum computing hackathon
https://qhack.ai
88 stars 124 forks source link

Predicting ground state of molecule with novel quantum descriptor using QML #121

Open justids opened 2 years ago

justids commented 2 years ago

Team Name:

SexyQuantumGuys

Project Description:

Classical Neural Networks (NNs) have been used in various ways to predict the ground state of molecules [1]. In such NN-based schemes, atom-centered symmetry functions (usually referred to as ‘descriptors’) are employed to ensure the translational, rotational, and permutation invariance of the many-body system. Considering the quantum version of this approach, we aim to achieve higher accuracy in predicting the ground state of molecules in shorter time, using Quantum Hybrid Neural Networks based on novel quantum descriptors with learnable Hamiltonian.

Also if possible, we are trying to make a Hamiltonian for the Bond Order Potential(BOP). Bonding potential is the largest part of molecule's energy. So by making a nice bonding Hamiltonian, the energy prediction will be much better. The energy will be predicted with the QAOA Method.

Presentation:

*A hyperlink to an explanatory presentation of your team’s hackathon project in a non-technical form (e.g., video, blog post, jupyter notebook, website, slideshow, [etc.] (https://github.com/justids/SexyQuantumGuys/blob/main/QNNP/QNNP_Tutorial.ipynb)

Source code:

*A hyperlink to the final source code for your team's hackathon project https://github.com/justids/SexyQuantumGuys

Which challenges/prizes would you like to submit your project for?

[Amazon Braket Challenge], [IBM Qiskit Challenge],[Hybrid Algorithms Challenge],[QAOA Challenge],[Quantum Chemistry Challenge],[Simulation Challenge],[Young Scientist Challenge]