Quantum Characterization and Model Reduction for Fault-Tolerant Qubit Networks
Abstract
Understanding the conditions under which fault-tolerant quantum computation (FTQC) may be practically achievable in realistic noisy quantum devices remains an outstanding challenge for noisy intermediate-scale quantum (NISQ) technology and beyond. While significant theoretical advances have been made in the simplified setting where the effective errors at the circuit level may be well-described as sufficiently local and Markovian, realistic scenarios typically involve noise sources that are both spatially and temporally correlated. Properly capturing the resulting correlated non-Markovian error behavior requires moving beyond the circuit-level picture, and reformulating the problem in terms of a Hamiltonian, open-quantum system picture. A main goal of the proposed research is to bridge the conceptual gulf that currently exists between circuit-level errors and Hamiltonian-level noise characteristics needed assess the viability for FTQC. This calls for the development of i) a unified theoretical framework able to connect the description of realistic spatio-temporally correlated noise sources at the physical level with their effective manifestations at the circuit level, along with ii) a comprehensive suite of algorithmic tools capable of efficiently characterizing noise features of relevance to FTQC. Solutions to these problems must be resource-efficient and extensible, as eventually they need to be applicable to large-scale multi-qubit networks comprising more than 50 qubits to verify and validate FTQC. The above challenges will be tackled by bringing together a diverse set of techniques from open quantum system and quantum control theory, statistical signal processing and quantum fault-tolerance. In particular, a key objective will be the development of systematic approaches for obtaining model-reduced descriptions for the controlled open-quantum system dynamics of large qubit networks, which have tractable complexity yet retain predictive power relative to desired performance metrics and can recover existing Markovian settings in well-defined limits. Technically, this will leverage the mathematical theory of frames as a general flexible framework to capture and prioritize the role of limited control and measurement capabilities, and will combine control-driven and NISQ-inspired data-driven model reduction methods. The team we have assembled for this project brings together three leading researchers, with complementary expertise and a demonstrated track record of impactful collaboration on relevant topics across quantum control engineering for quantum characterization, verification, and validation. Altogether, this endeavor will substantially advance our ability to efficiently characterize and operate large qubit networks exposed to realistic noise processes, and ultimately inform their potential for FTQC. In parallel, this project incorporates a strong educational and training component in theoretical quantum information science at both the postdoctoral and graduate level, thereby being directly responsive to the challenge of creating a quantum-smart workforce.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Feb 24, 2022
- Source ID
- W911NF2210004
Entities
People
- Lorenza Viola
Organizations
- Army Contracting Command
- Dartmouth College
- United States Army