Variational Optimization with Trapped-Ion Quantum Hardware
Abstract
Quantum computers offer a means for solving some problems faster than any conventional(classical) computer. However, the difficulty in accessing the information hidden within a complexentangled quantum superposition restricts the class of problems amenable to quantum computation,and the full domain of problems that admit a quantum advantage remains unknown.A promising quantum algorithmic framework for optimization problems is given by variationalquantum-classical hybrid algorithms. These invoke a number of layers of tailored quantum interactionson N qubits with parameters that are classically optimized in a closed-loop fashion toextremize a cost function on the distribution of the measured output state of the quantum system.An example which has gained particular traction in the last couple years is the QuantumApproximate Optimization Algorithm (QAOA), which uses the evolution of a quantum systemvia a bang-bang protocol. However, in this and other frameworks, initial experimental demonstrationsshow performance limitations due to errors in the gates or Hamiltonians implemented,or the large search space of variational parameters.The research proposed herein will pioneer hybrid quantum-classical optimization algorithms thatovercome the challenges associated with QAOA by exploiting unique capabilities of trappedions, one of the most advanced quantum computing and simulation platforms, together withnew quantum algorithms. The optimization problem is encoded in the native Ising-like spinHamiltonian of the trapped ion system, which is highly reconfigurable. We will use an approachthat relies on analog quantum simulation of individual steps in the algorithms, which allowseasier scaling to larger systems.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Aug 31, 2020
- Source ID
- N000142012695
Entities
People
- Guido Pagano
Organizations
- Office of Naval Research
- Rice University
- United States Navy