Mayar Elfares

Institute for Visualisation and Interactive Systems, University of Stuttgart
Institute of Information Security, University of Stuttgart

  Pfaffenwaldring 5a, 70569 Stuttgart, Germany University of Stuttgart, SimTech Building, Room 01.025 Google Scholar

Biography

Mayar Elfares is a PhD student in the Perceptual User Interface group since April 2022, jointly supervised with Prof. Ralf Küsters from the Institute of Information Security and funded by the University of Stuttgart AI Software Academy (AISA). She studied for her Master's and Bachelor's degrees at the German University in Cairo, Egypt, Department of Computer Science and Engineering. Her research interests are in Human-Computer Interaction, Computer Vision, and Privacy. She is conducting her Ph.D. in "privacy-preserving attentive user interface" at the Institute for Visualization and Interactive Systems as well as the Institute of Information Security.

Teaching

2022

Machine Perception and Learning (Tutor)Master

Open Thesis Projects

 

Federated Learning for Non-IID Gaze EstimationBachelor/Master

 

Private Mouse and Keyboard Behaviour DatasetBachelor

Publications

1. 

   Federated Learning for Appearance-based Gaze Estimation in the Wild

   Mayar Elfares, Zhiming Hu, Pascal Reisert, Andreas Bulling, Ralf Küsters

   Proceedings of the NeurIPS Workshop Gaze Meets ML (GMML), pp. 1–17, 2022.

   Abstract Links BibTeX Project

   Gaze estimation methods have significantly matured in recent years but the large number of eye images required to train deep learning models poses significant privacy risks. In addition, the heterogeneous data distribution across different users can significantly hinder the training process. In this work, we propose the first federated learning approach for gaze estimation to preserve the privacy of gaze data. We further employ pseudo-gradients optimisation to adapt our federated learning approach to the divergent model updates to address the heterogeneous nature of in-the-wild gaze data in collaborative setups. We evaluate our approach on a real-world dataset (MPIIGaze dataset) and show that our work enhances the privacy guarantees of conventional appearance-based gaze estimation methods, handles the convergence issues of gaze estimators, and significantly outperforms vanilla federated learning by 15.8% (from a mean error of 10.63 degrees to 8.95 degrees). As such, our work paves the way to develop privacy-aware collaborative 14 learning setups for gaze estimation while maintaining the model’s performance.

   doi: 10.48550/arXiv.2211.07330

   Paper: elfares22_gmml.pdf

   @inproceedings{elfares22_gmml, title = {Federated Learning for Appearance-based Gaze Estimation in the Wild}, author = {Elfares, Mayar and Hu, Zhiming and Reisert, Pascal and Bulling, Andreas and Küsters, Ralf}, year = {2022}, booktitle = {Proceedings of the NeurIPS Workshop Gaze Meets ML (GMML)}, doi = {10.48550/arXiv.2211.07330}, pages = {1--17} }

1. Multi-limb Split Learning for Tumor Classification on Vertically Distributed Data

   Omar Ads, Mayar Elfares, Mohammed Salem

   Proceedings of the Tenth International Conference on Intelligent Computing and Information Systems (ICICIS), pp. 88-92, 2021.

   Abstract BibTeX Project

   Brain tumors are one of the life-threatening forms of cancer. Previous studies have classified brain tumors using deep neural networks. In this paper, we perform the later task using a collaborative deep learning technique, more specifically split learning. Split learning allows collaborative learning via neural networks splitting into two (or more) parts, a client-side network and a server-side network. The client-side is trained to a specific layer named as the cut layer. Then, the rest of the training is resumed on the server-side network. Vertical distribution, a method for distributing data among organizations, was implemented where several hospitals hold different attributes of information for the same set of patients. To the best of our knowledge, this paper is the first paper to implement both split learning and vertical distribution for brain tumor classification. Using both techniques, we were able to achieve train and test accuracy greater than 90% and 70%, respectively.

   @inproceedings{ads21_icicis, author = {Ads, Omar and Elfares, Mayar and Salem, Mohammed}, title = {Multi-limb Split Learning for Tumor Classification on Vertically Distributed Data}, booktitle = {Proceedings of the Tenth International Conference on Intelligent Computing and Information Systems (ICICIS)}, year = {2021}, pages = {88-92} }