Machine Learning-Augmented Spectroscopies for Intelligent Materials Design

The thesis combines state-of-the-art materials characterization techniques with machine learning to extract information from neutron and photon spectroscopies. It proposes novel machine learning architectures to identify the proximity effect at topological material interfaces, which could enable spintronics and quantum computing with fault tolerance. The work sheds light on future pathways to apply machine learning to augment experiments.

Format: Hardback
Length: 97 pages
Publication date: 07 October 2022
Publisher: Springer International Publishing AG

The thesis makes groundbreaking contributions at the nexus of state-of-the-art materials characterization techniques and machine learning. By harnessing the power of machine learning, neutron and photon spectroscopies are empowered to extract more information, offering valuable insights into spectroscopic analysis for next-generation quantum technology. One notable example is the elusive proximity effect at topological material interfaces, which holds the potential to enable spintronics without energy dissipation and fault-tolerant quantum computing. However, identifying the characteristic spectral features of the proximity effect has been a challenge. The work presented in this thesis enables a precise resolution of its spectroscopic features and facilitates the determination of the proximity effect, which could enhance future experiments with improved interpretability. Additionally, the thesis proposes novel machine learning architectures that capitalize on scarce data and leverage the internal symmetry of systems to enhance training quality. This research sheds light on potential avenues for applying machine learning to augment experiments in the future.

Weight: 342g
Dimension: 235 x 155 (mm)
ISBN-13: 9783031148071
Edition number: 1st ed. 2022