Search Results - "Suriyakumar, Vinith"
-
1
Algorithms that Approximate Data Removal: New Results and Limitations
Published 25-09-2022“…We study the problem of deleting user data from machine learning models trained using empirical risk minimization. Our focus is on learning algorithms which…”
Get full text
Journal Article -
2
Algorithmic Pluralism: A Structural Approach To Equal Opportunity
Published 15-05-2024“…We present a structural approach toward achieving equal opportunity in systems of algorithmic decision-making called algorithmic pluralism. Algorithmic…”
Get full text
Journal Article -
3
When Personalization Harms: Reconsidering the Use of Group Attributes in Prediction
Published 04-06-2022“…Machine learning models are often personalized with categorical attributes that are protected, sensitive, self-reported, or costly to acquire. In this work, we…”
Get full text
Journal Article -
4
Unstable Unlearning: The Hidden Risk of Concept Resurgence in Diffusion Models
Published 10-10-2024“…Text-to-image diffusion models rely on massive, web-scale datasets. Training them from scratch is computationally expensive, and as a result, developers often…”
Get full text
Journal Article -
5
Architecture-Level Modeling of Photonic Deep Neural Network Accelerators
Published 14-05-2024“…ISPASS 2024 pp. 307-309 Photonics is a promising technology to accelerate Deep Neural Networks as it can use optical interconnects to reduce data movement…”
Get full text
Journal Article -
6
Architecture-Level Modeling of Photonic Deep Neural Network Accelerators
Published in 2024 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS) (05-05-2024)“…Photonics is a promising technology to accelerate Deep Neural Networks as it can use optical interconnects to reduce data movement energy and it enables…”
Get full text
Conference Proceeding -
7
One-shot Empirical Privacy Estimation for Federated Learning
Published 06-02-2023“…Privacy estimation techniques for differentially private (DP) algorithms are useful for comparing against analytical bounds, or to empirically measure privacy…”
Get full text
Journal Article -
8
Chasing Your Long Tails: Differentially Private Prediction in Health Care Settings
Published 13-10-2020“…Machine learning models in health care are often deployed in settings where it is important to protect patient privacy. In such settings, methods for…”
Get full text
Journal Article -
9
3D Reasoning for Unsupervised Anomaly Detection in Pediatric WbMRI
Published 24-03-2021“…Modern deep unsupervised learning methods have shown great promise for detecting diseases across a variety of medical imaging modalities. While previous…”
Get full text
Journal Article -
10
Private Multi-Winner Voting for Machine Learning
Published 23-11-2022“…Private multi-winner voting is the task of revealing $k$-hot binary vectors satisfying a bounded differential privacy (DP) guarantee. This task has been…”
Get full text
Journal Article -
11
Public Data-Assisted Mirror Descent for Private Model Training
Published 30-11-2021“…In this paper, we revisit the problem of using in-distribution public data to improve the privacy/utility trade-offs for differentially private (DP) model…”
Get full text
Journal Article -
12
Using Generative Models for Pediatric wbMRI
Published 01-06-2020“…Early detection of cancer is key to a good prognosis and requires frequent testing, especially in pediatrics. Whole-body magnetic resonance imaging (wbMRI) is…”
Get full text
Journal Article