Master's Defence
Shape and cutoff in superconducting qubits, work fluctuations in correlation creation, and critical commentary
Master's Candidate: Emma McKay
Master's Candidate: Emma McKay
Machine learning is an interesting family of problems for which near-term quantum devices can provide considerable advantages. In particular, exponential quantum speedup is recently demonstrated in learning a Boolean function that calculates the parity of a randomly chosen input bit string and a hidden bit string in the presence of noise, the problem known as learning parity with noise (LPN).
The quantum query complexity of a function f measures how many bits of the input a quantum computer must look at in order to compute f.
Observations reveal the cosmos to be astonishingly simple, and yet deeply puzzling, on the largest accessible scales. Why is it so nearly symmetrical? Why is there a cosmological constant (or dark energy) and what fixes its value? How did everything we see emerge from a singular “point” in the past?
Wavelength selective thermal emitters are highly desired for the development of the compact/energy-efficient spectroscopic sensing systems capable of detecting various gases such as COx, CH4, and NOx, which are strongly needed in environmental science, medical care, and other industrial applications. In addition, for the latter applications, dynamic control of thermal emission intensity is important for such emitters because synchronous detection can increase the signal-to-noise ratio significantly.
Quantum random-access memories (qRAM) are required by numerous quantum algorithms. In many cases, qRAM queries are the limiting factor in the implementation of these algorithms. In the limit of a large number of queries, there will be a massive resource overhead, as in this regime it is not possible to bypass the need for active error correction. In this talk, I will present our work towards quantifying this overhead. We will explore a variety of different qRAM circuits designed to query classical bits in superposition.
A random walk on a graph, P, with marked vertex set M, finds a marked vertex using a O(HT(P,M)) steps of the walk, where HT(P,M) is the hitting time. Previous quantum algorithms could detect the presence of a marked vertex in O(sqrt{HT(P,M)}) steps, or find a marked vertex in O(sqrt{HT(P,M)}) steps if M contained at most one vertex, but the case of finding in the presence of multiple marked vertices was left as an open problem.
Quantum (von Neumann) entropy optimization problems constitute an important class of quantum relative entropy optimization and have applications in various areas including quantum state tomography, statistical physics and machine learning. The optimization involves a nonlinear convex objective function $Tr(XlnX)$ and equality constraints on positive definite matrix $X$. In this talk I will present a long-step interior-point algorithm for the quantum entropy optimization (joint work with my co-advisor).
Rangefinding has many applications in navigation, civil engineer, construction, military, surveillance and security. Most commonly rangefinders estimate the distance to an object by measuring the time of flight of light for the journey to and returning from the target. Conventional techniques use lasers for illumination in state of the art rangefinding systems. However, the particular state of light lasers produce makes them easy to detect.
Master's Candidate: Maria Papageorgiou
Much of the structure of quantum field theory (QFT) is predicated on the principle of locality. Adherence to locality is pursuant to convictions deduced from relativity, and is achieved in QFT by the association of regions of spacetime with algebras of observables. Although, by construction, the observables of QFT are local objects, one may also consider characterizing the spatial or spacetime features of a state.