← Authors

Giulio Biroli

13 papers · 2018–2025 · 4 conferences · across top CS/AI conferences

Achievements

🌈 Renaissance Researcher (5) 🗺️ Taxonomy Completionist (23) 🌉 Interdisciplinary Bridge 🐣 Hot Topic Early Bird 🌍 Conference Polyglot (4) 🏃 Academic Marathon (7) 🧭 Keyword Pioneer 🐝 Cross-Pollinator (12) 🏆 Keyword Champion (2) 🔥 Unstoppable (5) 💎 Century Club (13) ❓ The Questioner (2) 🗃️ Keyword Collector (57)

Conferences

NIPS (7) ICML (4) AISTATS (1) ICLR (1)

Top co-authors

Levent Sagun (6) Chiara Cammarota (3) Florent Krzakala (3) Stéphane d'Ascoli (3) Stefano Sarao Mannelli (2) Stéphane D’Ascoli (2) Lenka Zdeborová (2) Franco Pellegrini (2) Hugo Touvron (1) Marylou Gabrie (1)

Keywords

neural network (4) gradient flow (3) double descent (2) shallow neural network (2) loss landscape (2) non-convex optimization (2) statistical physics (2) mean-field theory (2) interpolation threshold (2) phase retrieval (1) logistic loss (1) signal recovery (1) neural network optimization (1) generative model (1) singular value decomposition (1) generalization error (1) feature learning (1) inductive bia (1) convolutional neural network (1) gradient descent (1)

Papers

A Differentiable Rank-Based Objective for Better Feature Learning ICLR 2025
Cascade of phase transitions in the training of energy-based models NIPS 2024
On the Impact of Overparameterization on the Training of a Shallow Neural Network in High Dimensions AISTATS 2024
Neural Network Pruning Denoises the Features and Makes Local Connectivity Emerge in Visual Tasks ICML 2022
ConViT: Improving Vision Transformers with Soft Convolutional Inductive Biases ICML 2021
On the interplay between data structure and loss function in classification problems NIPS 2021
Double Trouble in Double Descent: Bias and Variance(s) in the Lazy Regime ICML 2020
Complex Dynamics in Simple Neural Networks: Understanding Gradient Flow in Phase Retrieval NIPS 2020
Triple descent and the two kinds of overfitting: where & why do they appear? NIPS 2020
An analytic theory of shallow networks dynamics for hinge loss classification NIPS 2020
Finding the Needle in the Haystack with Convolutions: on the benefits of architectural bias NIPS 2019
Who is Afraid of Big Bad Minima? Analysis of gradient-flow in spiked matrix-tensor models NIPS 2019
Comparing Dynamics: Deep Neural Networks versus Glassy Systems ICML 2018