Samuel Horvath

About me:

I am currently an assistant professor of Machine Learning at Mohamed bin Zayed University of Artificial Intelligence (MBZUAI). Prior to that, I completed my MS and Ph. D. in statistics at King Abdullah University of Science and Technology (KAUST), advised by professor Peter Richtárik. Before that, I was an undergraduate student in Financial Mathematics at Comenius University.

In my research, I focus on providing a fundamental understanding of how distributed and federated training algorithms work and how they interact with different sources of heterogeneity, such as system-level variability in the computing infrastructure and statistical variability in the training data. Inspired by the theoretical insights, I aim to design efficient and practical distributed/federated training algorithms.

I am broadly interested in federated learning, distributed optimization, and efficient deep learning.
I am always looking for collaborators. Please reach out if any of my work sounds interesting. For prospective students and interns, please have a look here.

Latest News (All News):

• (31/10/2022) Travelled to KAUST to attend and give a talk at the KAUST 2022 Workshop on Statistics

• (26/10/2022) New paper out

• (25/10/2022) Gave a talk on Federated Learning at Arab Youth Center

• (13/10/2022) Attended and gave a talk at the 1st CASL Workshop

• (12/10/2022) Attended GITEX Global to represent MBZUAI, gave a talk on Federated Learning and recorded a podcast.

• (08/10/2022) Co-organized Workshop on Collaborative Learning From Theory to Practice

• (26/09/2022) Paper accepted to TMLR

• (16/09/2022) Paper accepted to Nature: Communications Chemistry

• (05/09/2022) New Postdoc: Eduard Gorbunov joined MBZUAI, and he will be working with Martin Takac and me

• (22/08/2022) Paper accepted to TMLR

Experience:

• (Fall 2021) Facebook AI Research, Montreal, hosted by Mike Rabbat

• (Fall 2020) Samsung AI Center, Cambridge, hosted by Stefanos Laskaridis and Nicholas Lane

• (Summer 2019) Amazon AWS, Berlin, hosted by Cédric Archambeau

• (Spring 2019) University of California, Berkeley, hosted by Michael Jordan