58 citations · 129 across the 13 of their papers we have counts for
9 papers · 1 filter
Integral Probability Boundaries
Bogdan Kulynych, Shahab Asoodeh
Bounds based on Integral Probability Metrics (IPMs) are ubiquitous in machine learning and statistics, appearing, e.g., in generalization, alignment, fairness, and privacy. IPMs re…
Gaussian DP for Reporting Differential Privacy Guarantees in Machine Learning
Juan Felipe Gomez, Bogdan Kulynych, Georgios Kaissis +4
Current practices for reporting differential privacy (DP) guarantees for machine learning (ML) algorithms such as DP-SGD provide an incomplete and potentially misleading picture. F…
Attack-Aware Noise Calibration for Differential Privacy
Bogdan Kulynych, Juan Felipe Gomez, Georgios Kaissis +2
Differential privacy (DP) is a widely used approach for mitigating privacy risks when training machine learning models on sensitive data. DP mechanisms add noise during training to…
Lost in the Averages: A New Specific Setup to Evaluate Membership Inference Attacks Against Machine Learning Models
Nataša Krčo, Florent Guépin, Matthieu Meeus +2
Synthetic data generators and machine learning models can memorize their training data, posing privacy concerns. Membership inference attacks (MIAs) are a standard method of estima…
The Fundamental Limits of Least-Privilege Learning
Theresa Stadler, Bogdan Kulynych, Michael C. Gastpar +2
The promise of least-privilege learning -- to find feature representations that are useful for a learning task but prevent inference of any sensitive information unrelated to this…
Arbitrary Decisions are a Hidden Cost of Differentially Private Training
Bogdan Kulynych, Hsiang Hsu, Carmela Troncoso +1
Mechanisms used in privacy-preserving machine learning often aim to guarantee differential privacy (DP) during model training. Practical DP-ensuring training methods use randomizat…
Adversarial Robustness for Tabular Data through Cost and Utility Awareness
Klim Kireev, Bogdan Kulynych, Carmela Troncoso
Many safety-critical applications of machine learning, such as fraud or abuse detection, use data in tabular domains. Adversarial examples can be particularly damaging for these ap…
Disparate Vulnerability to Membership Inference Attacks
Bogdan Kulynych, Mohammad Yaghini, Giovanni Cherubin +2
A membership inference attack (MIA) against a machine-learning model enables an attacker to determine whether a given data record was part of the model's training data or not. In t…
Evading classifiers in discrete domains with provable optimality guarantees
Bogdan Kulynych, Jamie Hayes, Nikita Samarin +1
Machine-learning models for security-critical applications such as bot, malware, or spam detection, operate in constrained discrete domains. These applications would benefit from h…