Picture for Navid Azizan

Navid Azizan

Identifying Reliable Predictions in Detection Transformers

Add code
Dec 02, 2024
Viaarxiv icon

Optimizing Attention with Mirror Descent: Generalized Max-Margin Token Selection

Add code
Oct 18, 2024
Viaarxiv icon

Hard-Constrained Neural Networks with Universal Approximation Guarantees

Add code
Oct 14, 2024
Viaarxiv icon

Adapting Differentially Private Synthetic Data to Relational Databases

Add code
May 29, 2024
Viaarxiv icon

Private Synthetic Data Meets Ensemble Learning

Add code
Oct 15, 2023
Viaarxiv icon

A Unified Approach to Controlling Implicit Regularization via Mirror Descent

Add code
Jun 24, 2023
Figure 1 for A Unified Approach to Controlling Implicit Regularization via Mirror Descent
Figure 2 for A Unified Approach to Controlling Implicit Regularization via Mirror Descent
Figure 3 for A Unified Approach to Controlling Implicit Regularization via Mirror Descent
Figure 4 for A Unified Approach to Controlling Implicit Regularization via Mirror Descent
Viaarxiv icon

Representation Reliability and Its Impact on Downstream Tasks

Add code
May 31, 2023
Viaarxiv icon

SketchOGD: Memory-Efficient Continual Learning

Add code
May 25, 2023
Viaarxiv icon

On the Effects of Data Heterogeneity on the Convergence Rates of Distributed Linear System Solvers

Add code
Apr 20, 2023
Viaarxiv icon

Automatic Gradient Descent: Deep Learning without Hyperparameters

Add code
Apr 11, 2023
Viaarxiv icon