Alert button
Picture for Tolga Ergen

Tolga Ergen

Alert button

A Library of Mirrors: Deep Neural Nets in Low Dimensions are Convex Lasso Models with Reflection Features

Add code
Bookmark button
Alert button
Mar 02, 2024
Emi Zeger, Yifei Wang, Aaron Mishkin, Tolga Ergen, Emmanuel Candès, Mert Pilanci

Viaarxiv icon

The Convex Landscape of Neural Networks: Characterizing Global Optima and Stationary Points via Lasso Models

Add code
Bookmark button
Alert button
Dec 19, 2023
Tolga Ergen, Mert Pilanci

Viaarxiv icon

Fixing the NTK: From Neural Network Linearizations to Exact Convex Programs

Add code
Bookmark button
Alert button
Sep 26, 2023
Rajat Vadiraj Dwaraknath, Tolga Ergen, Mert Pilanci

Viaarxiv icon

Globally Optimal Training of Neural Networks with Threshold Activation Functions

Add code
Bookmark button
Alert button
Mar 06, 2023
Tolga Ergen, Halil Ibrahim Gulluk, Jonathan Lacotte, Mert Pilanci

Figure 1 for Globally Optimal Training of Neural Networks with Threshold Activation Functions
Figure 2 for Globally Optimal Training of Neural Networks with Threshold Activation Functions
Figure 3 for Globally Optimal Training of Neural Networks with Threshold Activation Functions
Figure 4 for Globally Optimal Training of Neural Networks with Threshold Activation Functions
Viaarxiv icon

Convexifying Transformers: Improving optimization and understanding of transformer networks

Add code
Bookmark button
Alert button
Nov 20, 2022
Tolga Ergen, Behnam Neyshabur, Harsh Mehta

Figure 1 for Convexifying Transformers: Improving optimization and understanding of transformer networks
Figure 2 for Convexifying Transformers: Improving optimization and understanding of transformer networks
Figure 3 for Convexifying Transformers: Improving optimization and understanding of transformer networks
Figure 4 for Convexifying Transformers: Improving optimization and understanding of transformer networks
Viaarxiv icon

GLEAM: Greedy Learning for Large-Scale Accelerated MRI Reconstruction

Add code
Bookmark button
Alert button
Jul 18, 2022
Batu Ozturkler, Arda Sahiner, Tolga Ergen, Arjun D Desai, Christopher M Sandino, Shreyas Vasanawala, John M Pauly, Morteza Mardani, Mert Pilanci

Figure 1 for GLEAM: Greedy Learning for Large-Scale Accelerated MRI Reconstruction
Figure 2 for GLEAM: Greedy Learning for Large-Scale Accelerated MRI Reconstruction
Figure 3 for GLEAM: Greedy Learning for Large-Scale Accelerated MRI Reconstruction
Figure 4 for GLEAM: Greedy Learning for Large-Scale Accelerated MRI Reconstruction
Viaarxiv icon

Unraveling Attention via Convex Duality: Analysis and Interpretations of Vision Transformers

Add code
Bookmark button
Alert button
May 20, 2022
Arda Sahiner, Tolga Ergen, Batu Ozturkler, John Pauly, Morteza Mardani, Mert Pilanci

Figure 1 for Unraveling Attention via Convex Duality: Analysis and Interpretations of Vision Transformers
Figure 2 for Unraveling Attention via Convex Duality: Analysis and Interpretations of Vision Transformers
Figure 3 for Unraveling Attention via Convex Duality: Analysis and Interpretations of Vision Transformers
Figure 4 for Unraveling Attention via Convex Duality: Analysis and Interpretations of Vision Transformers
Viaarxiv icon

Path Regularization: A Convexity and Sparsity Inducing Regularization for Parallel ReLU Networks

Add code
Bookmark button
Alert button
Oct 25, 2021
Tolga Ergen, Mert Pilanci

Figure 1 for Path Regularization: A Convexity and Sparsity Inducing Regularization for Parallel ReLU Networks
Figure 2 for Path Regularization: A Convexity and Sparsity Inducing Regularization for Parallel ReLU Networks
Figure 3 for Path Regularization: A Convexity and Sparsity Inducing Regularization for Parallel ReLU Networks
Figure 4 for Path Regularization: A Convexity and Sparsity Inducing Regularization for Parallel ReLU Networks
Viaarxiv icon