Eugene Belilovsky

Eugene Belilovsky is an Assistant Professor in the Department of Computer Science and Software Engineering at Concordia University. His research focuses on machine learning, computer vision, optimization algorithms, and their applications in medical imaging and federated learning. He explores topics such as continual learning, model merging, and bias mitigation in deep learning systems. His work bridges theoretical advancements with practical applications, including healthcare diagnostics and efficient training strategies for large models.

Belilovsky's recent publications emphasize innovative methods like MuLoCo for inner optimizer design, FairDropout for enhancing minority group generalization, and techniques to scale multi-task learning with sparse masks. His studies on federated learning explore incentive mechanisms for decentralized systems and robust pre-training of large language models. His contributions to 3D reconstruction and motion prediction highlight interdisciplinary applications in computer graphics and generative models.

His research trends reflect a strong emphasis on improving model efficiency, fairness, and adaptability across diverse domains. While no scientific awards are explicitly listed, his prolific publication record underscores his impactful contributions to AI and machine learning.