Picture for Hassan Ashtiani

Hassan Ashtiani

Agnostic Private Density Estimation via Stable List Decoding

Add code
Jul 05, 2024
Viaarxiv icon

Sample-Optimal Locally Private Hypothesis Selection and the Provable Benefits of Interactivity

Add code
Dec 09, 2023
Viaarxiv icon

Mixtures of Gaussians are Privately Learnable with a Polynomial Number of Samples

Add code
Sep 07, 2023
Viaarxiv icon

On the Role of Noise in the Sample Complexity of Learning Recurrent Neural Networks: Exponential Gaps for Long Sequences

Add code
May 28, 2023
Viaarxiv icon

Polynomial Time and Private Learning of Unbounded Gaussian Mixture Models

Add code
Mar 07, 2023
Viaarxiv icon

Benefits of Additive Noise in Composing Classes with Bounded Capacity

Add code
Jun 14, 2022
Figure 1 for Benefits of Additive Noise in Composing Classes with Bounded Capacity
Figure 2 for Benefits of Additive Noise in Composing Classes with Bounded Capacity
Figure 3 for Benefits of Additive Noise in Composing Classes with Bounded Capacity
Figure 4 for Benefits of Additive Noise in Composing Classes with Bounded Capacity
Viaarxiv icon

Adversarially Robust Learning with Tolerance

Add code
Mar 02, 2022
Viaarxiv icon

Private and polynomial time algorithms for learning Gaussians and beyond

Add code
Nov 22, 2021
Viaarxiv icon

Privately Learning Mixtures of Axis-Aligned Gaussians

Add code
Jun 03, 2021
Viaarxiv icon

On the Sample Complexity of Privately Learning Unbounded High-Dimensional Gaussians

Add code
Oct 19, 2020
Viaarxiv icon