Picture for Mohamad Amin Mohamadi

Mohamad Amin Mohamadi

Adam Exploits $\ell_\infty$-geometry of Loss Landscape via Coordinate-wise Adaptivity

Add code
Oct 10, 2024
Viaarxiv icon

Why Do You Grok? A Theoretical Analysis of Grokking Modular Addition

Add code
Jul 17, 2024
Viaarxiv icon

Making Look-Ahead Active Learning Strategies Feasible with Neural Tangent Kernels

Add code
Jun 25, 2022
Figure 1 for Making Look-Ahead Active Learning Strategies Feasible with Neural Tangent Kernels
Figure 2 for Making Look-Ahead Active Learning Strategies Feasible with Neural Tangent Kernels
Figure 3 for Making Look-Ahead Active Learning Strategies Feasible with Neural Tangent Kernels
Figure 4 for Making Look-Ahead Active Learning Strategies Feasible with Neural Tangent Kernels
Viaarxiv icon

A Fast, Well-Founded Approximation to the Empirical Neural Tangent Kernel

Add code
Jun 25, 2022
Figure 1 for A Fast, Well-Founded Approximation to the Empirical Neural Tangent Kernel
Figure 2 for A Fast, Well-Founded Approximation to the Empirical Neural Tangent Kernel
Figure 3 for A Fast, Well-Founded Approximation to the Empirical Neural Tangent Kernel
Figure 4 for A Fast, Well-Founded Approximation to the Empirical Neural Tangent Kernel
Viaarxiv icon