Scientific Calendar Event

An ICTP Hybrid meeting

“Youth in high-dimensions” gives to young researchers from academia and industry the opportunity to gather and present recent results on high-dimensional statistical problems, in areas ranging from machine learning and inference to neuroscience or statistical physics.
Typical problems in high-dimensions include estimating noisy signals (compressed sensing, PCA and tensor decomposition), analysing (deep) neural-networks, detecting communities in large networks, or modelling disordered spin systems. These seemingly unrelated problems share many similarities in their phenomenology and in the tools used to analyse them. The core of the field is made of a very active, diverse and quickly expanding community of physicists, computer scientists, mathematicians, information theorists and engineers, with the common desire to tackle increasingly challenging problems at the forefront of data science.

This conference aims at reinforcing the links among this interdisciplinary community, and in particular of its youngest theory-oriented members, and to bring forward the latest development happening in the high-dimensional world.
  • High-dimensional statistics
  • Statistical physics
  • Machine learning
  • Computational neuroscience
  • Complex systems
  • Machine learning for science
  • Statistical inference

M. ALBERGO, New York University
K. CHANDRASEKHER, Stanford University
R. DUDEJA, Harvard University
S. FREI, UC Berkeley
R. GHEISSARI, Northwestern
S. GOEL, University of Pennsylvania
A. JALAL, UC Berkeley
E. MATHIEU, University of Cambridge
S. PAPPALARDI, University of Köln
P. ROTONDO, University of Parma

Applicants are encouraged to submit an abstract for a poster presentation. During the application, please make sure to use our templates to format your abstract. Templates are available below for download.
Grants: A limited number of grants are available to support the attendance of selected participants, with priority given to participants from developing countries. There is no registration fee.
Go to day
  • Monday, 29 May 2023
    • 08:30 - 09:00 Registration and administrative formalities
      REGISTRATION: Upon arrival, Visitors not staying in the ICTP Guest Houses, are kindly requested to complete registration formalities at the Adriatico Guest House (Lower level). The Registration Desk will be open from 8.00 to 8.50.
    • 08:50 - 14:00
      • 08:50 Opening remarks 10'
        Speakers: Marylou Therese GABRIE (École Polytechnique), Sebastian GOLDT (SISSA), Marco MONDELLI (ISTA), Jean BARBIER (ICTP)
      • 09:00 Implicit regularization and benign overfitting for neural networks in high dimensions 45'
        Speaker: S. FREI (UC Berkeley)
        Material: Abstract Slides Video
      • 09:45 Convergence guarantees for iterative algorithms in a class of nonconvex empirical risk minimization problems 45'
        Speaker: K. CHANDRASEKHER (Stanford University)
        Material: Abstract Video
      • 10:30 Coffee break 30'
      • 11:00 Stochastic Interpolants: A unifying framework for flows and diffusions 45'
        Speaker: M. ALBERGO (New York University)
        Material: Abstract Slides Video
      • 11:45 Geometric Neural Diffusion Processe 45'
        Speaker: E. MATHIEU (University of Cambridge)
        Material: Abstract Slides Video
      • 12:30 Lunch 1h30'
    • 14:00 - 18:15 Quantum session
      Chair: V. ROS (Laboratoire de Physique Théorique et Modèles Statistiques)
      • 14:00 Introduction to Deep Reinforcement Learning with applications in Quantum Control 45'
        Speaker: M. BUKOV (MPIKS Dresden)
        Material: Abstract Video
      • 14:45 How Free Probability appears in mesoscopic quantum systems 45'
        Speaker: L. HRUZA (ENS Paris)
        Material: Abstract Slides Video
      • 15:30 The Eigenstate Thermalization Hypothesis and Free Probability 45'
        Speaker: S. PAPPALARDI (University of Köln)
        Material: Abstract Video
      • 16:15 Coffee break 30'
      • 16:45 Poster session 1h30'
        In person
        Material: Booklet Poster Abstracts
  • Tuesday, 30 May 2023
    • 09:00 - 20:55
      • 09:00 High-dimensional Limit Theorems for Stochastic Gradient Descent 45'
        Speaker: R. GHEISSARI (Northwestern)
        Material: Abstract Video
      • 09:45 Finite-Width Neural Networks: A Landscape Complexity Analysis 45'
        Speaker: B. SIMSEK (EPFL)
        Material: Abstract Video
      • 10:30 Coffee break 30'
      • 11:00 Discussion 1h30'
      • 12:30 Lunch 1h30'
      • 14:00 Matrix Factorization with Neural Networks of Associative Memory 45'
        Speaker: F. CAMILLI (ICTP)
        Material: Abstract Slides Video
      • 14:45 Coffee break 15'
      • 15:00 Spectral Universality in Regularized Linear Regression 45'
        Speaker: R. DUDEJA (Harvard University)
        Material: Abstract Video
      • 15:45 Break 15'
      • 16:00 HECAP-MATH Special Lecture: How will we do mathematics in 2030 ? 1h30'
        see: LINK to Overview page
        Speaker: Michael DOUGLAS (Stony Brook University and Harvard University, USA)
        Material: Video
      • 17:40 Hike & Welcome reception 3h15'
  • Wednesday, 31 May 2023
    • 09:00 - 17:30
      • 09:00 Learning hierarchical compositionality with deep convolutional networks: insights from a Random Hierarchy Model 45'
        Speaker: F. CAGNETTA (EPFL)
        Material: Abstract Video
      • 09:45 Statistical mechanics of deep learning beyond the infinite-width limit 45'
        Speaker: P. ROTONDO (University of Parma)
        Material: Abstract Slides Video
      • 10:30 Coffee break 30'
      • 11:00 Discussion 1h30'
      • 12:30 Lunch 1h30'
      • 14:00 Compressed Sensing using Generative Models: Theory and Applications 45'
        Speaker: A. JALAL (UC Berkeley)
        Material: Abstract Video
      • 14:45 Machine-learning-assisted Monte Carlo fails at sampling computationally hard problems 45'
        Speaker: J. TRINQUIER (ENS)
        Material: Abstract Slides Video
      • 15:30 Group photo & Coffee break 30'
        Online & 'In person'
      • 16:00 Discussion 1h30'
  • Thursday, 1 June 2023
    • 09:00 - 14:00
      • 09:00 Thinking fast with Transformers - Algorithmic Reasoning via Shortcuts 45'
        Speaker: S. GOEL (University of Pennsylvania)
        Material: Abstract Video
      • 09:45 Discovering spikes in random matrices using approximate message passing 45'
        Speaker: Y. ZHANG (ISTA)
        Material: Abstract Slides Video
      • 10:30 Coffee break 30'
      • 11:00 The staircase property and the leap complexity 45'
        Speaker: E. BOIX-ADSERA (MIT)
        Material: Abstract Video
      • 11:45 Data-centric AI: Play with Datasets to Improve Machine Learning Performances 45'
        Speaker: F. GERACE (SISSA)
        Material: Abstract Video
      • 12:30 Lunch 1h30'
    • 19:00 - 23:00 ICTP Social Event
      Location: Leonardo Building Terrace
      Material: Poster
  • Friday, 2 June 2023
    • 09:00 - 14:00
      • 09:00 Discussion & closing 3h0'
      • 12:00 Lunch 2h0'