Anthony Fernando Judeson

About Me

I am a recent graduate with a dual Master’s degree in Electrical Engineering and Information Technology from the Technical University of Munich (TUM, Germany) and the École Nationale Supérieure de l’Électronique et de ses Applications (ENSEA, France). My academic journey began with digital systems, signal processing, and embedded systems, before progressively evolving toward deep learning and computer vision.

I further strengthened this specialization during an exchange at the Korea Advanced Institute of Science and Technology (KAIST, South Korea), and through my Master’s thesis at BMW Group in Munich, where I worked on computer vision and model optimization for autonomous driving applications.

Research Interests

I am driven by the challenge of designing deep neural networks that are both high-performing and computationally efficient for deployment in real-world scenarios. Balancing accuracy with resource constraints is especially critical for embedded systems and edge devices.

My research focuses on compression techniques such as quantization, pruning, and structural reparameterization—leveraging feature map and latent space analysis to streamline inference without compromising accuracy.

I am also interested in advancing network robustness, particularly in building models resilient to adversarial attacks and capable of continual learning without access to the original training dataset. These issues are essential in contexts with strict constraints on privacy, storage, and computation.

Beyond robustness and efficiency, I am drawn to 3D vision, diffusion models, and generative approaches, which hold strong potential for applications in medical imaging and autonomous systems.

Publications & Research Projects