AI Center Seminar - AI Fundamentals series - Prof. Riccardo Zecchina

The talk is organized by the EPFL AI Center as part of the AI fundamentals seminar series. Talk followed by a coffee session.

Hosting professor: Prof. Lenka Zdeborová

Title
Dynamical Learning in Deep Asymmetric Recurrent Neural Networks

Abstract
We show that asymmetric deep recurrent neural networks, enhanced with additional sparse excitatory couplings, give rise to an exponentially large, dense accessible manifold of internal representations which can be found by different algorithms, including simple iterative dynamics. Building on the geometrical properties of the stable configurations, we propose a distributed learning scheme in which input-output associations emerge naturally from the recurrent dynamics, without any need of gradient evaluation. A critical feature enabling the learning process is the stability of the configurations reached at convergence, even after removal of the supervisory output signal. Extensive simulations demonstrate that this approach performs competitively on standard AI benchmarks. The model can be generalized in multiple directions, both computational and biological, potentially contributing to narrowing the gap between AI and computational neuroscience.

Bio
Riccardo Zecchina (RZ) is Professor of Theoretical Physics at Bocconi University in Milan, where he holds the Chair in Machine Learning. His research spans the intersection of statistical physics, computer science, advanced machine learning, and computational neuroscience.

He earned a degree in Electronic Engineering from the Polytechnic University of Turin, followed by a PhD in Theoretical Physics at the University of Turin under the supervision of Tullio Regge. From 1997 to 2007, he was a researcher and head of the Statistical Physics group at the International Centre for Theoretical Physics (ICTP) in Trieste. He later served as Full Professor of Theoretical Physics at the Polytechnic University of Turin (2007–2017). In 2017, he joined Bocconi University, where he established the Department of Computing Sciences and developed new degree programs in mathematical and computational methods for Artificial Intelligence.

Professor Zecchina has held long-term visiting positions at Microsoft Research (Redmond and Cambridge, MA) and at the Laboratory of Theoretical Physics and Statistical Models (LPTMS) at the University of Paris-Sud. His contributions to the field have been recognized with major awards, including the Lars Onsager Prize in Theoretical Statistical Physics from the American Physical Society (2016, jointly with M. Mézard and G. Parisi). He was also awarded an ERC Advanced Grant from the European Research Council (2011–2015).