conftrace_

initialization scale

4 papers

Explore in graph

Co-occurring keywords · papers mentioning both

implicit bia (2) gradient flow (2) gradient descent (2) monarch matrice (1) block tensor-train (btt) (1) compute-efficient models (1) memorization tasks (1) deep linear classification (1) kernel regime (1) structured matrice (1)

Papers

An Analysis for Reasoning Bias of Language Models with Small Initialization ICML 2025 Compute Better Spent: Replacing Dense Layers with Structured Matrices ICML 2024 On the Implicit Bias of Initialization Shape: Beyond Infinitesimal Mirror Descent ICML 2021 Implicit Bias in Deep Linear Classification: Initialization Scale vs Training Accuracy NeurIPS 2020