Picture for Can Yaras

Can Yaras

Compressible Dynamics in Deep Overparameterized Low-Rank Learning & Adaptation

Jun 06, 2024
Viaarxiv icon

Understanding Deep Representation Learning via Layerwise Feature Compression and Discrimination

Add code
Nov 06, 2023
Viaarxiv icon

The Law of Parsimony in Gradient Descent for Learning Deep Linear Networks

Add code
Jun 01, 2023
Figure 1 for The Law of Parsimony in Gradient Descent for Learning Deep Linear Networks
Figure 2 for The Law of Parsimony in Gradient Descent for Learning Deep Linear Networks
Figure 3 for The Law of Parsimony in Gradient Descent for Learning Deep Linear Networks
Figure 4 for The Law of Parsimony in Gradient Descent for Learning Deep Linear Networks
Viaarxiv icon

Neural Collapse with Normalized Features: A Geometric Analysis over the Riemannian Manifold

Add code
Sep 19, 2022
Figure 1 for Neural Collapse with Normalized Features: A Geometric Analysis over the Riemannian Manifold
Figure 2 for Neural Collapse with Normalized Features: A Geometric Analysis over the Riemannian Manifold
Figure 3 for Neural Collapse with Normalized Features: A Geometric Analysis over the Riemannian Manifold
Figure 4 for Neural Collapse with Normalized Features: A Geometric Analysis over the Riemannian Manifold
Viaarxiv icon

Randomized Histogram Matching: A Simple Augmentation for Unsupervised Domain Adaptation in Overhead Imagery

Apr 30, 2021
Figure 1 for Randomized Histogram Matching: A Simple Augmentation for Unsupervised Domain Adaptation in Overhead Imagery
Figure 2 for Randomized Histogram Matching: A Simple Augmentation for Unsupervised Domain Adaptation in Overhead Imagery
Figure 3 for Randomized Histogram Matching: A Simple Augmentation for Unsupervised Domain Adaptation in Overhead Imagery
Figure 4 for Randomized Histogram Matching: A Simple Augmentation for Unsupervised Domain Adaptation in Overhead Imagery
Viaarxiv icon