Jeremias Sulam

Dr. Jeremias Sulam is an Assistant Professor in the Biomedical Engineering Department at Johns Hopkins University's Whiting School of Engineering, where he has been faculty since 2018. His research bridges theoretical machine learning with biomedical applications, focusing on interpretability, adversarial robustness, and representation learning.

Dr. Sulam's research interests center on making machine learning models more interpretable and robust, with particular emphasis on:

• Formal frameworks for sufficient and necessary explanations in model interpretability
• Proximal operators and novel approaches to diffusion models
• Robustness certification against sparse adversarial perturbations
• Sparse representations and dictionary learning for inverse problems
• Concept bottleneck models that maintain performance while increasing interpretability

His recent publications (2024-2025) show a strong focus on explainable AI and generative models, with theoretical contributions that have practical implications for medical imaging and other biomedical applications. Dr. Sulam's work consistently combines mathematical rigor with practical implementation, as evidenced by his publications in top machine learning venues.

Dr. Sulam's research is supported by active collaborations with researchers across multiple institutions, as shown by his diverse co-author list spanning computer science, mathematics, and biomedical engineering disciplines. His work demonstrates how theoretical machine learning advances can be applied to solve real-world problems in healthcare and biomedical domains.