conftrace_

Salma Tarmoun

4 papers · 2021–2025 · 2 conferences · across top CS/AI conferences

Achievements

Jump to papers ↓

🌍 Conference Polyglot (2) 🌉 Interdisciplinary Bridge 🧭 Keyword Pioneer 🐣 Hot Topic Early Bird 🐝 Cross-Pollinator (15)

Conferences

AISTATS (2) ICML (2)

Top co-authors

René Vidal (4) Hancheng Min (3) Enrique Mallada (3) Ziqing Xu (2) Guilherme Franca (1) Benjamin D Haeffele (1) Jinqi Luo (1) Lachlan Ewen MacDonald (1)

Keywords

linear network (2) gradient descent (2) gradient flow (1) implicit bia (1) overparameterized model (1) linear model (1) riccati equation (1) neural network (1) matrix factorization (1) min-norm solution (1) convergence rate (1)

Papers

Understanding the Learning Dynamics of LoRA: A Gradient Flow Perspective on Low-Rank Adaptation in Matrix Factorization AISTATS 2025 Linear Convergence of Gradient Descent For Finite Width Over-parametrized Linear Networks With General Initialization AISTATS 2023 On the Explicit Role of Initialization on the Convergence and Implicit Bias of Overparametrized Linear Networks ICML 2021 Understanding the Dynamics of Gradient Flow in Overparameterized Linear models ICML 2021