Thesis
Predicting the von Neumann entanglement entropy using a graph neural network
University of Iowa
Master of Science (MS), University of Iowa
Summer 2025
DOI: 10.25820/etd.008062
Abstract
Calculating the von Neumann entanglement entropy from experimental data is challenging due to its dependence on the complete wavefunction, forcing reliance on approximations such as classical mutual information (MI). We propose a machine learning approach using a graph neural network to predict the von Neumann entropy directly from experimentally accessible bitstrings. We test this approach on an Ising model configurations and a Rydberg ladder system and achieve a mean absolute error of 2.6 × 10−3 and 3.6 × 10−3 respectively when evaluating within the training range on a dataset with entropy values ranging from 0 to 1.9. The models achieves a mean absolute percentage error of 1.49% and 1.44% and outperforms MI-based bounds. When tested beyond the training range, the models maintains reasonable accuracy. Furthermore, we demonstrate for the Rydberg system that fine-tuning the model with small datasets significantly improves performance on data outside the original training range.
Details
- Title: Subtitle
- Predicting the von Neumann entanglement entropy using a graph neural network
- Creators
- Anas Saleh
- Contributors
- Yannick Meurice (Advisor)Wayne N Polyzou (Committee Member)Craig E Pryor (Committee Member)
- Resource Type
- Thesis
- Degree Awarded
- Master of Science (MS), University of Iowa
- Degree in
- Physics
- Date degree season
- Summer 2025
- DOI
- 10.25820/etd.008062
- Publisher
- University of Iowa
- Number of pages
- x, 44 pages
- Copyright
- Copyright 2025 Anas Saleh
- Grant note
This work was supported in part by the Department of Energy under Award Number DE-SC0010113.
(ii)- Comment
- This thesis has been optimized for improved web viewing. If you require the original version, contact the University Archives at the University of Iowa: https://www.lib.uiowa.edu/sc/contact/
- Language
- English
- Date submitted
- 07/28/2025
- Description illustrations
- illustrations, tables, graphs
- Description bibliographic
- Includes bibliographical references (pages 40-44).
- Public Abstract (ETD)
Quantum systems can become entangled, meaning their particles are mysteriously connected even when separated. Scientists want to measure this entanglement, but it s extremely difficult with current experimental tools. This research develops a new method using artificial intelligence to predict entanglement levels from data that experiments can easily collect. We trained computer models on two different quantum systems and found they could accurately predict entanglement with less than 2% error. This new approach works better than existing methods and could help scientists better understand and control quantum systems for future technologies like quantum computers.
- Academic Unit
- Physics and Astronomy
- Record Identifier
- 9984948237902771
Metrics
2 File views/ downloads
9 Record Views