About
I’m a Master’s student in Computer Science at the Data Systems Group, within the Cheriton School of Computer Science at the University of Waterloo, supervised by Prof. Xi He. I’m also a graduate researcher at the Vector Institute and expect to graduate in January 2027.
Since May 2026, I’ve been working as a Software Developer Intern (AI/ML) at Ford Motor Company on conversational and offline AI assistants for vehicle infotainment, with an expected end date of December 2026. My work includes embedded language models, speech-to-intent recognition, and model optimization for limited compute.
Before starting my Master’s, I worked as a research assistant and completed an internship at the Institute for Research in Fundamental Sciences (IPM), under the supervision of Prof. Dara Rahmati. I earned my Bachelor’s degree in Computer Science from Shahid Beheshti University in 2023.
My research focuses on generative modeling, synthetic data, and privacy in multi-relational tabular systems. I have also worked on LLM agent inference, federated learning for medical imaging, multimodal emotion recognition, and autonomous-driving perception. Beyond research, I enjoy building practical AI systems, from real-time anomaly detection to Android apps and recommender systems.
I’m always excited to connect, exchange ideas, and collaborate on meaningful projects—feel free to reach out!
Publications
FERMI: Exploiting Relations for Membership Inference Against Tabular Diffusion Models, Abtin Mahyar, Masoumeh Shafieinejad, Yuhan Liu, Xi He, ICML 2026 (FoGen). OpenReview
Feature extraction and deep learning, Abtin Mahyar, Hossein Motamednia, Pooryaa Cheraaqee, Azadeh Mansouri, Dimensionality Reduction in Machine Learning, pp. 211-243. Morgan Kaufmann.
Cultural-Aware AI Model for Emotion Recognition, Mehrdad Baradaran, Payam Zohari, Abtin Mahyar, Hossein Motamednia, Dara Rahmati, Saeid Gorgin, IEEE International Machine Vision and Image Processing Conference (MVIP), 2024. Code
Brain age estimation using structural MRI: a clustered federated learning approach, Seyyed Saeid Cheshmi, Abtin Mahyar, Anita Soroush, Zahra Rezvani, Bahar Farahani, IEEE International Conference on Omni-layer Intelligent Systems (COINS), 2023. Code