I am a Master's student in Electrical and Computer Engineering at Georgia Tech. Previously, I was a research intern in the Physics Laboratory at ENS de Lyon, France, where I worked with Pierre Borgnat and Paulo Gonçalves on machine learning techniques for neuroscience, specifically focusing on detecting Seizure Onset Zones (SOZ) from stereo-EEG signals in epilepsy patients. This work resulted in an accepted extended abstract at the Graph Signal Processing Workshop (MILA, Montréal, May 2025), and I am currently working to publish a full paper based on this research.

Email / CV / LinkedIn

Research

My research focuses on developing more effective and interpretable representation learning methods, particularly for high-dimensional biomedical signals and multimodal data. I work at the intersection of contrastive learning, Transformer architectures, and domain-specific knowledge integration to solve real-world problems.

A key contribution of my work is the development of spatial contrastive pre-training strategies for Transformer encoders applied to stereo-EEG signal analysis. This approach leverages the spatial relationships between electrodes to improve seizure onset zone detection, demonstrating how domain knowledge can enhance representation learning in clinical applications.

My research has spanned multiple domains: from adapting Transformer architectures for computer vision tasks (as demonstrated in my work on historical document conservation) to applying Graph Neural Networks for representation learning.

Leveraging clinically-informed time-frequency features and spatial contrastive pre-training within a Transformer encoder for improved Seizure Onset Zone (SOZ) localization from stereo-EEG.

Developed and evaluated a Vision Transformer-based system for multi-label classification of defects on historical book bindings to aid conservation efforts.