Join Nir Shlezinger, Assistant Professor in the School of Electrical and Computer Engineering in Ben-Gurion University, Israel, and Yonina Eldar, Professor in the Department of Mathematics and Computer Science, Weizmann Institute of Science, for a discussion about designing machine learning systems that are inherently efficient.

By drawing inspiration from classical algorithms, we can create machine learning models that are computationally efficient, interpretable, and robust—this brings us to the concept of model-based deep learning. Model-based deep learning integrates traditional signal processing principles with modern data-driven methods. It allows us to achieve efficiency not just through computational optimizations, but by structuring learning algorithms in a way that inherently aligns with physical models and mathematical structures.

This episode explores the key aspects of model-based deep learning, its benefits, and its applications in signal processing.

Click here to listen to the experts or watch on YouTube.