Print Email Facebook Twitter Improving machine learning based side-channel analysis: can dropout be dropped out? Title Improving machine learning based side-channel analysis: can dropout be dropped out? Author Kok, Jim (TU Delft Electrical Engineering, Mathematics and Computer Science) Contributor Picek, S. (mentor) Krcek, M. (mentor) Degree granting institution Delft University of Technology Programme Computer Science and Engineering Project CSE3000 Research Project Date 2020-06-22 Abstract Analysing physical leakages (e.g. power consumption and electromagnetic radiation) of cryptographic devices can be used by adversaries to extract secret keys. Over the last couple of years, researchers have shown that machine learning has potential for this process. Machine learning models need to be fine-tuned to enhance key extraction performance. This paper investigates the so-called dropout hyper-parameter, which is proven to reduce overfitting in various domains (e.g. speech recognition). Dropout is examined for two different models: multilayer perceptrons and convolutional neural networks. Regarding the convolutional neural networks, two architectures are examined: one architecture used as a benchmark in various papers and a more uncomplicated one. The findings of this paper showed that adding dropout to the investigated multilayer perceptron architecture led to significant improvements, whereas the convolutional neural network architectures showed negligible improvements. Subject Side-channel analysismachine learningoverfittingdropout To reference this document use: http://resolver.tudelft.nl/uuid:3391941a-6910-436e-98c5-c7c5b6ddd4fd Part of collection Student theses Document type bachelor thesis Rights © 2020 Jim Kok Files PDF Research_project_final_paper.pdf 497.49 KB Close viewer /islandora/object/uuid:3391941a-6910-436e-98c5-c7c5b6ddd4fd/datastream/OBJ/view