Florent Krzakala

Full Professor@Swiss Federal Institute of Technology Lausanne·瑞士·OpenReview

研究方向

Statistical Physics · High-dimensional asymptotics · Universality · Statistical learning theory · Dynamics of SGD · Replica Method · Approximate message Passing · Multi-index · teacher-student · asymptotics · SGD · Information Exponent

7.0

Scaling Laws and Spectra of Shallow Neural Networks in the Feature Learning Regime

ICLR 2026Oral

通讯

3.5

Inductive Bias and Spectral Properties of Single-Head Attention in High Dimensions

ICLR 2026Rejected

7.8

The Computational Advantage of Depth in Learning High-Dimensional Hierarchical Targets

NeurIPS 2025Spotlight

通讯

7.6

Learning with Restricted Boltzmann Machines: Asymptotics of AMP and GD in High Dimensions

NeurIPS 2025Poster

二作

6.8

Optimal Spectral Transitions in High-Dimensional Multi-Index Models

NeurIPS 2025Poster

6.8

Asymptotics of SGD in Sequence-Single Index Models and Single-Layer Attention Networks

NeurIPS 2025Poster

6.4

The Nuclear Route: Sharp Asymptotics of ERM in Overparameterized Quadratic Networks

NeurIPS 2025Poster

通讯

5.5

Fundamental limits of learning in sequence multi-index models and deep attention networks: high-dimensional asymptotics and sharp thresholds

合作者 (19)