1 citations · 3 across the 6 of their papers we have counts for
6 papers
Inference Attacks for X-Vector Speaker Anonymization
Luke Bauer, Wenxuan Bao, Malvika Jadhav +1
We revisit the privacy-utility tradeoff of x-vector speaker anonymization. Existing approaches quantify privacy through training complex speaker verification or identification mode…
Provably Secure Covert Messaging Using Image-based Diffusion Processes
Luke A. Bauer, Wenxuan Bao, Vincent Bindschaedler
We consider the problem of securely and robustly embedding covert messages into an image-based diffusion model's output. The sender and receiver want to exchange the maximum amount…
On the Importance of Architecture and Feature Selection in Differentially Private Machine Learning
Wenxuan Bao, Luke A. Bauer, Vincent Bindschaedler
We study a pitfall in the typical workflow for differentially private machine learning. The use of differentially private learning algorithms in a "drop-in" fashion -- without acco…
Attacks as Defenses: Designing Robust Audio CAPTCHAs Using Attacks on Automatic Speech Recognition Systems
Hadi Abdullah, Aditya Karlekar, Saurabh Prasad +5
Audio CAPTCHAs are supposed to provide a strong defense for online resources; however, advances in speech-to-text mechanisms have rendered these defenses ineffective. Audio CAPTCHA…
Leveraging Generative Models for Covert Messaging: Challenges and Tradeoffs for "Dead-Drop" Deployments
Luke A. Bauer, James K. Howes, Sam A. Markelon +2
State of the art generative models of human-produced content are the focus of many recent papers that explore their use for steganographic communication. In particular, generative…
Generative Models for Security: Attacks, Defenses, and Opportunities
Luke A. Bauer, Vincent Bindschaedler
Generative models learn the distribution of data from a sample dataset and can then generate new data instances. Recent advances in deep learning has brought forth improvements in…