Degree
Master of Science in Data Science
Department
Department of Computer Science
Faculty/ School
School of Mathematics and Computer Science (SMCS)
Date of Submission
Spring 2023
Supervisor
Dr. Jibran Rashid, Assistant Professor, Department of Computer Science, School of Mathematics and Computer Science (SMCS)
Committee Member 1
Dr. Tahir Syed, Examiner – I, Institute of Business Administration (IBA), Karachi
Committee Member 2
Dr. Imran Rauf, Examiner-II, Institute of Business Administration
Keywords
Nonlocal Correlations, Quantum Entanglement, No-Signaling Polytope, High-Dimensional Polytopes, Classical Polytope, Bell’s Theorem, Bell Inequalities, Quantum Polytope
Abstract
Nonlocal correlations are a quantum phenomenon shown to have a stronger form of correlation than classical correlations (Bell 1964), (Rohrlich 1994), (B. S. Tsirelson 1980). Nonlocal correlations are generated using quantum entanglement. Nonlocal correlations live in an exponentially large space and hence are computationally hard to classify. Research in the past century has shown quantum entanglement and nonlocal correlations to be a crucial feature of quantum computation and information.
Recently machine learning techniques have been used to classify the space of nonlocal correlations. However, the literature on the subject is either scarce (Askery Canabarro 2018), or favors deep neural networks (Valcarce 2018). Supervised learning techniques requiring less resources and a single layer algorithm are capable of classifying these correlations at a comparable accuracy.
In this thesis we initiate building a classifier between signaling and no-signaling nonlocal correlations as well as classical vs. non-classical correlations. The current literature on the subject also takes a multi-algorithm approach to classifying nonlocal correlations. Our work has taken a much simpler approach which attempts to classify these correlations using a single algorithm.
If we are successful in defining clear boundaries between classical, no-signaling, and signaling regions using machine learning algorithms, we can build on this approach to attempt to classify quantum and non-quantum correlations; a task made more difficult due to the lack of inequalities to define the quantum region in the probability space.
Document Type
Restricted Access
Submission Type
Thesis
Recommended Citation
Mehmood, M. Y. (2023). Application of Machine Learning Techniques for classification of Nonlocal Correlations (Unpublished graduate thesis). Institute of Business Administration, Pakistan. Retrieved from https://ir.iba.edu.pk/research-projects-msds/16
The full text of this document is only accessible to authorized users.
