activity
20242026
collaborators

5 papers

math.OC2026

Phases of Muon: When Muon Eclipses SignSGD

Elliot Paquette, Noah Marshall, Lucas Benigni +3

Recently, Muon and related spectral optimizers have demonstrated strong empirical performance as scalable stochastic methods, often outperforming Adam. Yet their behaviour remains…

stat.ML2026

Logarithmic-time Schedules for Scaling Language Models with Momentum

Damien Ferbach, Courtney Paquette, Gauthier Gidel +2

In practice, the hyperparameters and weight-decay in AdamW are typically kept at fixed values. Is there any reason to do otherwise? We show that for large-scale…

stat.ML2025

Dimension-adapted Momentum Outscales SGD

Damien Ferbach, Katie Everett, Gauthier Gidel +2

We investigate scaling laws for stochastic momentum algorithms with small batch on the power law random features model, parameterized by data complexity, target complexity, and mod…

stat.ML2025

4+3 Phases of Compute-Optimal Neural Scaling Laws

Elliot Paquette, Courtney Paquette, Lechao Xiao +1

We consider the solvable neural scaling model with three parameters: data complexity, target complexity, and model-parameter-count. We use this neural scaling model to derive new p…

math.OC2024

The High Line: Exact Risk and Learning Rate Curves of Stochastic Adaptive Learning Rate Algorithms

Elizabeth Collins-Woodfin, Inbar Seroussi, Begoña García Malaxechebarría +3

We develop a framework for analyzing the training and learning rate dynamics on a large class of high-dimensional optimization problems, which we call the high line, trained using…