The doi above is for this specific version of this dataset, which is currently the latest. Newer versions may be published in the future. For a link that will always point to the latest version, please use
doi: 10.4121/fe14a263-d5f1-4d3e-8d06-b1be95904acf

Krček, Marina (2024): Code underlying the publication: Autoencoder-enabled model portability for reducing hyperparameter tuning efforts in side-channel analysis. Version 1. 4TU.ResearchData. software. https://doi.org/10.4121/fe14a263-d5f1-4d3e-8d06-b1be95904acf.v1

Software

• Other Information and Computing Sciences
• Information and Computing Sciences

Autoencoders, Hyperparameter Tuning, Portability, Preprocessing, Side-channel Analysis, Transfer Learning

CC BY 4.0

• Data collection for PhD thesis: Implementation Attacks Powered by Artificial Intelligence

RefWorks, BibTeX, Reference Manager, Endnote, DataCite, NLM, DC, CFF

by Marina Krček

Link to GitHub repository with source code for the publication: Autoencoder-enabled model portability for reducing hyperparameter tuning efforts in side-channel analysis.

The source code uses the Python programming language. Scripts used to run the experiments are in the main directory, while the folder 'src' holds the implementations for hyperparameter tuning, loading of side-channel datasets, etc., providing some abstraction. Scripts starting with 'attack' were used to run experiments, while other scripts were helper scripts for analyzing/reading/plotting results.

Sbatch scripts were used to run experiments with TU Delft servers.

More information can be found in the publication.

• 2024-05-14 first online, published, posted

4TU.ResearchData

python scripts (.py), slurm sbatch scripts (.sbatch)

Autoencoder-enabled model portability for reducing hyperparameter tuning efforts in side-channel analysis

TU Delft, Faculty of Electrical Engineering, Mathematics and Computer Science, Department of Intelligent Systems