Biography
Bernardo Pulido-Gaytan received a Ph.D. degree in Computer Science from the CICESE Research Center, Mexico in 2024. His expertise lies in designing privacy-preserving machine learning cognitive models in cloud environments using homomorphic encryption cryptosystems. He has been involved in collaborative efforts with multiple international research groups, leading to several research internships at high-level institutions such as the University of Göttingen in Germany, the North-Caucasus Federal University in Russia, the University of the Republic in Uruguay, and the National High Technology Center in Costa Rica, among others. He has served as thesis supervisor for undergraduate and master's students, delivered invited talks at international conferences, and collaborated as a program committee member at several esteemed conferences. Bernardo is currently a Postdoctoral Research Associate at the Cloud Competency Center, National College of Ireland (NCI).
Research
My research is focused on comprehensively investigating the design of privacy-preserving Neural Network (NN) models for classifying encrypted information using lattice-based HE cryptosystems. The homomorphic processing of cognitive models, such as NNs, requires operations not supported by HE, so a main topic in the field is to find cryptographically compatible replacement non-linear functions to operate over ciphertexts.
Publications
Book chapters
- Pulido-Gaytan, Bernardo, Tchemykh, Andrei, Babenko, Mikhail, Cortés-Mendoza, Jorge M., González-Vélez, Horacio and Avetisyan, Arutyun (2024) "Enhancing Cloud Security through Efficient Polynomial Approximations for Homomorphic Evaluation of Neural Network Activation Functions". In: 2024 IEEE 24th International Symposium on Cluster, Cloud and Internet Computing Workshops (CCGridW). IEEE, Philadelphia, PA, USA, pp. 42-49. ISBN 979-8-3503-7751-4