The field of deep learning has seen dramatic developments over the last decade. Improved fundamental understanding of deep learning can enhance our ability to train deeper and more advanced...Show moreThe field of deep learning has seen dramatic developments over the last decade. Improved fundamental understanding of deep learning can enhance our ability to train deeper and more advanced networks. We have explored random neural networks. These networks exhibit two phases depending on the weight variances used to initialise them. Through numerical simulations, we are able to confirm the existence of these two phases and the associated emerging depth scales. Furthermore, we characterise the edge of chaos that separates these two phases. We find that significant differences between activation layer distributions on this edge of chaos exist. This raises the question as to whether some points on this edge show improved training performance over others.Show less
The coexistence of the semi-conducting 2H and quasi-metallic 1T’ monolayer MoS2 phases has great potential in, for example, low-resistance contacts. In this research MoS2 on gold samples have been...Show moreThe coexistence of the semi-conducting 2H and quasi-metallic 1T’ monolayer MoS2 phases has great potential in, for example, low-resistance contacts. In this research MoS2 on gold samples have been fabricated. Several different fabrication procedures have been tested including stamping, gold-mediated exfoliation and hydrogen plasma cleaning. Measurements identifying the 2H phase have been made using Atomic Force Microscopy and Low Energy Electron Microscopy. Although 1T’ could not be observed, insights are gained into the various fabrication processes.Show less
In this project, magneto-transport measurements are performed on exfoliated trigonal tantalum disulfide (1T-TaS2) flakes with a top-contact Hall-bar geometry. Transport measurements reveal the...Show moreIn this project, magneto-transport measurements are performed on exfoliated trigonal tantalum disulfide (1T-TaS2) flakes with a top-contact Hall-bar geometry. Transport measurements reveal the presence of charge density waves and therelatednearly-commensuratetocommensuratephasetransition. Thephase transition is shown to depend on both the thickness of the crystal and the cooling rate. The observed critical thickness is approximately 80 nm, relatively large compared to literature. A clear deviation from literature is observed in the resistive behavior during the transition. The increase in resistivity due to the transition is considerably smaller and less abrupt. This deviation may be attributed to partial switching of the crystal, i.e. only some of the layers switching to the commensurate phase. Hall effect measurements in the supercooled phase, i.e. below the suppressed phase transition, reveal large charge carrier density and extremely low mobility, both in agreement with literature. A down-turn is observed in the supercooled phase at low temperatures. This down-turn is also observed in literature, however its origin is not discussed. We propose that this down-turn is caused by weak anti-localization. The phase coherence length, extracted from fitting of the Hikami-Larkin-Nagaoka model (HLN)modeltotheweakanti-localizationpeaks,followsapowerlawwithexponent γ = −0.341±0.03. This exponent suggests that the electron transport is one-dimensional, substantiating the notion that in the nearly-commensurate phase, the electron transport is dominated by the domain boundaries.Show less
