arXiv subjects

Samuel Prescott

Publications and source records attributed to Samuel Prescott.

2 recordsLinked to original sources

arXiv2026

Simultaneous anti-bunched and super-bunched photons from a GaAs Quantum dot in a dielectric metasurface

Semiconductor quantum dots host a rich manifold of excitonic complexes, including neutral excitons that emit anti-bunched single photons and charged exciton complexes capable of producing super-bunched photons via cascade emission. Accessing both emission regimes from a single emitter would open routes to novel quantum protocols, including advanced quantum imaging. In practice, however, emission from charged exciton complexes is intrinsically weak, often orders of magnitude dimmer than neutral excitons, placing simultaneous dual-mode operation out of reach. Here, we overcome this limitation by embedding the quantum dot in a dielectric Mie-resonant metasurface that provides order-of-magnitude photoluminescence enhancement across both neutral and charged exciton transitions of a single GaAs quantum dot. Under identical non-resonant pumping conditions, the emission from the neutral exciton yields anti-bunched emission ($g^{(2)}(0) < 0.5$) and the emission from positively charged exciton complexes shows super-bunched emission ($g^{(2)}(0) > 3.5$) with comparable count rates (~12 kHz). Crucially, super-bunching emerges only when charged exciton emission spectrally overlaps with the Mie resonances and vanishes in un-patterned slabs, demonstrating that photonic engineering, is essential for accessing these weak quantum light states. These results demonstrate a scalable, position-tolerant platform for harnessing the full excitonic structure of solid-state emitters.

cond-mat.mes-hall↗

arXiv2023

An FPGA Architecture for Online Learning using the Tsetlin Machine

There is a need for machine learning models to evolve in unsupervised circumstances. New classifications may be introduced, unexpected faults may occur, or the initial dataset may be small compared to the data-points presented to the system during normal operation. Implementing such a system using neural networks involves significant mathematical complexity, which is a major issue in power-critical edge applications. This paper proposes a novel field-programmable gate-array infrastructure for online learning, implementing a low-complexity machine learning algorithm called the Tsetlin Machine. This infrastructure features a custom-designed architecture for run-time learning management, providing on-chip offline and online learning. Using this architecture, training can be carried out on-demand on the \ac{FPGA} with pre-classified data before inference takes place. Additionally, our architecture provisions online learning, where training can be interleaved with inference during operation. Tsetlin Machine (TM) training naturally descends to an optimum, with training also linked to a threshold hyper-parameter which is used to reduce the probability of issuing feedback as the TM becomes trained further. The proposed architecture is modular, allowing the data input source to be easily changed, whilst inbuilt cross-validation infrastructure allows for reliable and representative results during system testing. We present use cases for online learning using the proposed infrastructure and demonstrate the energy/performance/accuracy trade-offs.

cs.LG↗