86 citations · 201 across the 19 of their papers we have counts for
20 papers
Model Card for OpenAI Privacy Filter
Charles de Bourcy, Sahra Ghalebikesabi, Avi Schwarzschild +22
OpenAI Privacy Filter is a compact, bidirectional token-classification model for detecting and redacting personally identifiable information (PII) and secrets in unstructured text.…
DPrivBench: Benchmarking LLMs' Reasoning for Differential Privacy
Erchi Wang, Pengrun Huang, Eli Chien +4
Differential privacy (DP) has a wide range of applications for protecting data privacy, but designing and verifying DP algorithms requires expert-level reasoning, creating a high b…
Unintended Memorization in Large ASR Models, and How to Mitigate It
Lun Wang, Om Thakkar, Rajiv Mathews
It is well-known that neural networks can unintentionally memorize their training examples, causing privacy concerns. However, auditing memorization in large non-auto-regressive au…
Why Is Public Pretraining Necessary for Private Model Training?
Arun Ganesh, Mahdi Haghifam, Milad Nasr +5
In the privacy-utility tradeoff of a model trained on benchmark language and vision tasks, remarkable improvements have been widely reported with the use of pretraining on publicly…
Recycling Scraps: Improving Private Learning by Leveraging Intermediate Checkpoints
Virat Shejwalkar, Arun Ganesh, Rajiv Mathews +5
In this work, we focus on improving the accuracy-variance trade-off for state-of-the-art differentially private machine learning (DP ML) methods. First, we design a general framewo…
Measuring Forgetting of Memorized Training Examples
Matthew Jagielski, Om Thakkar, Florian Tramèr +8
Machine learning models exhibit two seemingly contradictory phenomena: training data memorization, and various forms of forgetting. In memorization, models overfit specific trainin…
Detecting Unintended Memorization in Language-Model-Fused ASR
W. Ronny Huang, Steve Chien, Om Thakkar +1
End-to-end (E2E) models are often being accompanied by language models (LMs) via shallow fusion for boosting their overall quality as well as recognition of rare words. At the same…
Extracting Targeted Training Data from ASR Models, and How to Mitigate It
Ehsan Amid, Om Thakkar, Arun Narayanan +2
Recent work has designed methods to demonstrate that model updates in ASR training can leak potentially sensitive attributes of the utterances used in computing the updates. In thi…
Revealing and Protecting Labels in Distributed Training
Trung Dang, Om Thakkar, Swaroop Ramaswamy +3
Distributed learning paradigms such as federated learning often involve transmission of model updates, or gradients, over a network, thereby avoiding transmission of private data.…
A Method to Reveal Speaker Identity in Distributed ASR Training, and How to Counter It
Trung Dang, Om Thakkar, Swaroop Ramaswamy +3
End-to-end Automatic Speech Recognition (ASR) models are commonly trained over spoken utterances using optimization methods like Stochastic Gradient Descent (SGD). In distributed s…
Practical and Private (Deep) Learning without Sampling or Shuffling
Peter Kairouz, Brendan McMahan, Shuang Song +3
We consider training models with differential privacy (DP) using mini-batch gradients. The existing state-of-the-art, Differentially Private Stochastic Gradient Descent (DP-SGD), r…
Training Production Language Models without Memorizing User Data
Swaroop Ramaswamy, Om Thakkar, Rajiv Mathews +3
This paper presents the first consumer-scale next-word prediction (NWP) model trained with Federated Learning (FL) while leveraging the Differentially Private Federated Averaging (…
Privacy Amplification via Random Check-Ins
Borja Balle, Peter Kairouz, H. Brendan McMahan +2
Differentially Private Stochastic Gradient Descent (DP-SGD) forms a fundamental building block in many applications for learning over sensitive data. Two standard approaches, priva…
Understanding Unintended Memorization in Federated Learning
Om Thakkar, Swaroop Ramaswamy, Rajiv Mathews +1
Recent works have shown that generative sequence models (e.g., language models) have a tendency to memorize rare or unique sequences in the training data. Since useful models are o…
Evading Curse of Dimensionality in Unconstrained Private GLMs via Private Gradient Descent
Shuang Song, Thomas Steinke, Om Thakkar +1
We revisit the well-studied problem of differentially private empirical risk minimization (ERM). We show that for unconstrained convex generalized linear models (GLMs), one can obt…
Guaranteed Validity for Empirical Approaches to Adaptive Data Analysis
Ryan Rogers, Aaron Roth, Adam Smith +3
We design a general framework for answering adaptive statistical queries that focuses on providing explicit confidence intervals along with point estimates. Prior work in this area…
Differentially Private Learning with Adaptive Clipping
Galen Andrew, Om Thakkar, H. Brendan McMahan +1
Existing approaches for training neural networks with user-level differential privacy (e.g., DP Federated Averaging) in federated learning (FL) settings involve bounding the contri…
Model-Agnostic Private Learning via Stability
Raef Bassily, Om Thakkar, Abhradeep Thakurta
We design differentially private learning algorithms that are agnostic to the learning model. Our algorithms are interactive in nature, i.e., instead of outputting a model based on…
Differentially Private Matrix Completion Revisited
Prateek Jain, Om Thakkar, Abhradeep Thakurta
We provide the first provably joint differentially private algorithm with formal utility guarantees for the problem of user-level privacy-preserving collaborative filtering. Our al…
Max-Information, Differential Privacy, and Post-Selection Hypothesis Testing
Ryan Rogers, Aaron Roth, Adam Smith +1
In this paper, we initiate a principled study of how the generalization properties of approximate differential privacy can be used to perform adaptive hypothesis testing, while giv…