Collection of notebooks for experiments in the use of (Quantum) Machine Learning for entanglement detection.
This repository is part of the project for the exam of Advanced Quantum Mechanics @ units 2023/24.
- Simone Brusatin <[email protected]>
- Christian Candeago <[email protected]>
- Alessio D'Anna <alessiomaria.d'[email protected]>
- Paolo Da Rold <[email protected]>
The following is a brief description of the subdirectories of the project.
More detailed information is contained in each subdirectory README.md.
datasets/Contains the datasets used for the project and the notebook to generate them.qsvm/Contains various notebooks that use Quantum Support Vector Machines to classify the data.qvc/Contains notebooks using Quantum Variational Circuits like Quantum Neural Networks, _HybridNNs_Classifiers and Quantum Convolutional NNs.classic/Contains notebook that use non-quantum ML techniques.images/Contains images for the readmes.Quantum_Machine_learning___Entanglement_detection_FINALESlides (in italian) of the presentation for the exam workshop.
The goal of the project is classify separable and entangled quantum states using (Quantum) Machine Learning techniques.
Entanglement is an essential resource for quantum information processing tasks, but finding a robust and efficient method for detecting entanglement is still an open problem. Various criterion exist to determine if a state is entangled, but these criterions require full quantum state tomography followed by density matrix estimation. In particular, we aim to create an entanglement approach which maximizes accuracy while minimizing the number of measurements of the system.
More detailed descriptions of each algorithm and the dataset generation are found in each subdirectories README.md.
Various datasets are generated that cover pure and mixed states. From the previous datasets, observables are applied to obtain new datasets. Various machine learning algorithms are then applied to classify the data.