Graph states in large-scale integrated quantum photonics

Abstract

The fragility of quantum states rapidly leads to errors in the operation of quantum computers. However, error-correction schemes overcome this challenge by entangling many physical qubits into composite logical qubits, which are error-protected. Consequently, any viable platform for quantum computing must demonstrate a route for the generation and control of large-scale entangled states. In the measurement-based model of quantum computing, error correction is intrinsically encoded in entangled resource states, known as graph states, such that the fault-tolerant operation of logical qubits naturally arises. Theoretical proposals for graph state processing in the platform of integrated photonics are appealing because, in principle, today’s foundry tools might be sufficient to fabricate the required large number of components. However, the crucial experimental steps of realising graph states, logical qubits
and error-correction schemes in an integrated photonics device have, until now, not been demonstrated.

In this thesis, we propose and demonstrate integrated photonic schemes for the realisation of large reconfigurable graph states based on qubit and qudit encodings. We implement this photonic architecture on programmable silicon chips that can generate high-quality graph states with up to eight qubits. Reporting an increase in both the number of on-chip generated photons and
their local dimension, we substantially expand the space of entanglement classes it is possible to experimentally realise. Embedding several error-correction encoding schemes into graph states, we explore measurement-based protocols and applications in both the physical and logical graph state scenarios, showing significant improvements in the computational performance of our
platform. Finally, we show the first experimental investigation of hypergraph states, generalised resources for novel approaches to measurement-based quantum computing, with potential for protecting against correlated errors.

These results are an important step forward for correcting quantum errors in CMOS compatible technologies. As research around the world focuses on developing applications for noisy intermediate-scale quantum devices, early noise-reduction methods will provide greater scope for running quantum algorithms. Subsequent progression towards fault-tolerant quantum computing
with integrated photonics can only be achieved by further development of the results presented in this thesis.

Date of Award	29 Sept 2020
Original language	English
Awarding Institution	University of Bristol
Supervisor	Anthony Laing (Supervisor) & John G Rarity (Supervisor)