Abstract:The development of large-scale platforms for quantum information requires new methods for verification and validation of quantum behavior. Quantum tomography (QT) is the standard tool for diagnosing quantum states, process, and readout devices by providing complete information. However, QT is limited since it is expensive to not only implement experimentally, but also requires heavy classical post-processing of data. In this dissertation, we introduce new methods for QT that are more efficient to implement and robust to noise and errors, thereby making QT a more practical tool for current quantum information experiments. The crucial detail that makes these new, efficient, and robust methods possible is prior information about the quantum system. This prior information is prompted by the goals of most experiments in quantum information, which require pure states, unitary processes, and rank-1 POVM operators. Therefore, most experiments are designed to operate near this ideal regime. We show that when this is the case, QT can be accomplished with significantly fewer resources, and produce a robust estimate in the presence of noise and errors. Moreover, the estimate is also robust if the state is not exactly pure, the process is not exactly unitary, or the POVM is not exactly rank-1. Such compelling methods are only made possible by the positivity constraint on quantum states, processes, and POVMs. This requirement is an inherent feature of quantum mechanics, but has powerful consequences to QT. Since QT is necessarily an experimental tool, we discuss a test of these new methods in an experimental setting. The physical system is an ensemble of laser-cooled cesium atoms in the laboratory of Prof. Poul Jessen. Experiments were conducted by Hector Sosa-Martinez et al. to demonstrate different QT protocols. We compare the results, and conclude that the new methods are effective.

Efficient Quantum State Tracking in Noisy Environments

Projected gradient descent algorithms for quantum state tomography

Learning quantum states of continuous variable systems

Fitting quantum noise models to tomography data

Efficient learning of arbitrary single-copy quantum states

Non-Markovian Noise Characterization with the Transfer Tensor Method

Efficient learning of mixed-state tomography for photonic quantum walk

Efficient factored gradient descent algorithm for quantum state tomography

Simple, reliable and noise-resilient continuous-variable quantum state tomography with convex optimization

Efficient and Robust Methods for Quantum Tomography

Matrix-Completion Quantum State Tomography

Quantum State Tomography with Locally Purified Density Operators and Local Measurements

Efficient Experimental Qudit State Estimation via Point Tomography

Efficient Quantum Mixed-State Tomography with Unsupervised Tensor Network Machine Learning

Fast and Noise-Robust Quantum State Tomography Based on ELM

Fast Quantum State Reconstruction via Accelerated Non-Convex Programming

Quantum State Tomography of Multi-Qubit Systems — a Comparative Study

Sample Optimal and Memory Efficient Quantum State Tomography

Experimental neural network enhanced quantum tomography

Experimental Demonstration of Efficient Quantum State Tomography of Matrix Product States