BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:AI Center Seminar - AI Fundamentals series - Prof. Riccardo Zecchi
 na
DTSTART:20251007T140000
DTEND:20251007T150000
DTSTAMP:20260427T142355Z
UID:51045c6eae4cf3fa9e2f86d9dda393eee6ac0b027d4c34071363e043
CATEGORIES:Conferences - Seminars
DESCRIPTION:Prof. Riccardo Zecchina\nThe talk is organized by the EPFL A
 I Center as part of the AI fundamentals seminar series. Talk followed by
  a coffee session.\n\nHosting professor: Prof. Lenka Zdeborová\n\nTitle\n
 Dynamical Learning in Deep Asymmetric Recurrent Neural Networks\n\nAbstrac
 t\nWe show that asymmetric deep recurrent neural networks\, enhanced with 
 additional sparse excitatory couplings\, give rise to an exponentially lar
 ge\, dense accessible manifold of internal representations which can be fo
 und by different algorithms\, including simple iterative dynamics. Buildin
 g on the geometrical properties of the stable configurations\, we propose 
 a distributed learning scheme in which input-output associations emerge na
 turally from the recurrent dynamics\, without any need of gradient evaluat
 ion. A critical feature enabling the learning process is the stability of 
 the configurations reached at convergence\, even after removal of the supe
 rvisory output signal. Extensive simulations demonstrate that this approac
 h performs competitively on standard AI benchmarks. The model can be gener
 alized in multiple directions\, both computational and biological\, potent
 ially contributing to narrowing the gap between AI and computational neuro
 science.\n\nBio\nRiccardo Zecchina (RZ) is Professor of Theoretical Physi
 cs at Bocconi University in Milan\, where he holds the Chair in Machine Le
 arning. His research spans the intersection of statistical physics\, compu
 ter science\, advanced machine learning\, and computational neuroscience.\
 n\nHe earned a degree in Electronic Engineering from the Polytechnic Unive
 rsity of Turin\, followed by a PhD in Theoretical Physics at the Universit
 y of Turin under the supervision of Tullio Regge. From 1997 to 2007\, he w
 as a researcher and head of the Statistical Physics group at the Internati
 onal Centre for Theoretical Physics (ICTP) in Trieste. He later served as 
 Full Professor of Theoretical Physics at the Polytechnic University of Tur
 in (2007–2017). In 2017\, he joined Bocconi University\, where he establ
 ished the Department of Computing Sciences and developed new degree progra
 ms in mathematical and computational methods for Artificial Intelligence.\
 n\nProfessor Zecchina has held long-term visiting positions at Microsoft R
 esearch (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\, inclu
 ding the Lars Onsager Prize in Theoretical Statistical Physics from the Am
 erican 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).
LOCATION:ELE 117 https://plan.epfl.ch/?room==ELE%20117 https://epfl.zoom.u
 s/j/63385346150?pwd=Fy9xm4XqBrSTaahwgdC3AkfSBRUKCq.1
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR