Characterization

Accurate noise mitigation requires one to fully understand what types of noise are present in systems. The characterization thrust will develop quantum characterization methods with an emphasis on robust, scalable, and platform independent methods that can be incorporated into automated routines.

Efficiency methods for multi-qubit Quantum Noise Spectroscopy

Scalable methods for quantum noise spectroscopy though encoded spaces

JHU/APL and Dartmouth

Bayesian Parameter Learning

Parameterized models together with Bayesian experimental design allow one to rapidly infer model

Lawrence Livermore National Lab

Gate-based Control

This thrust will develop quantum control methods at the circuit or gate level to improve error-resilience and robustness for near-term quantum algorithms. Emphasis will be places on hardware-agnostic general methods that can be used across platforms.

Algorithmic Error Mitigation

Addition of realistic noise into zero-noise extrapolation for improved performance

Unitary Fund

Pulse-Level Error Mitigation

Deep
Reinforcement
Learning

Filter shaping through DRL for rapid gate synthesis

JHU/APL and Darthmouth

Circuit-Level Error Mitigation

Correlations maintained in time through circuit level error model

JHU/APL and Darthmouth

Aggregated Control

This thrust will focus on developing optimal control methods for directly synthesizing large unitary operations into aggregated instructions to reduce circuit depth. This will enable fast, low-noise operation on NISQ quantum processors.

Aggregated Control

Multiple controls (top) for a circuit are aggregated into a much shorter set of controls (bottom) for improved noise resilience

Chicago and Lawrence Livermore National Lab

HPC Based Control Optimization & Scaling

20 qubits: 1.1 trillion DOF
Largest seismic simulation on 1200 nodes of summit: 2.5 trillion DOF + 126,000 timesteps
Large HPC-based routines should allow for multi-qubit control optimizations

Chicago and Lawrence Livermore National Lab

Characterization & Control

Three Step Strategy to Control-Based Noise Mitigation