Distributed Learning in a Quantum-Classical Hybrid System
Machine learning and deep learning have made significant advances in recent years, transforming the way we approach and solve complex problems. However, the viability of machine learning and deep learning on classical computers has began to come into question as the need for complex problem solving has increased since classical computers are limited in terms of their processsing power. Quantum computing has emerged as a new form of computing that uses quantum mechanics to process and manipulate information. Quantum computers use quantum bits, or qubits, which can exist in multiple states simultaneously. This property allows quantum computers to perform computations more efficiently than classical computers. However, publicly available quantum resources are limited to five or seven qubits per machine which limits the complexity of the problems that can be solved. A potential solution to resource limitation challenges is the use of a distributed system which is a network of interconnected computers that work together to achieve a common goal. In this thesis, a distributed quantum system is introduced. Our system allows for the efficient management of qubits on individual machines while expanding the qubits available by managing a distributed system of quantum computers. This solution will lead to the ability to solve increasingly complex problems while maximizing the use of limited quantum resources.
Computer science|Artificial intelligence|Information science
D'Onofrio, Anthony, "Distributed Learning in a Quantum-Classical Hybrid System" (2023). ETD Collection for Fordham University. AAI30488199.