Search Results for author: Bogdan Kulynych

Found 13 papers, 6 papers with code

The Fundamental Limits of Least-Privilege Learning

no code implementations19 Feb 2024 Theresa Stadler, Bogdan Kulynych, Nicoals Papernot, Michael Gastpar, Carmela Troncoso

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 task -- is highly appealing.

Attribute

Prediction without Preclusion: Recourse Verification with Reachable Sets

no code implementations24 Aug 2023 Avni Kothari, Bogdan Kulynych, Tsui-Wei Weng, Berk Ustun

In turn, models can assign predictions that are fixed, meaning that consumers who are denied loans, interviews, or benefits may be permanently locked out from access to credit, employment, or assistance.

Adversarial Robustness

Arbitrary Decisions are a Hidden Cost of Differentially Private Training

1 code implementation28 Feb 2023 Bogdan Kulynych, Hsiang Hsu, Carmela Troncoso, Flavio P. Calmon

We demonstrate that such randomization incurs predictive multiplicity: for a given input example, the output predicted by equally-private models depends on the randomness used in training.

Privacy Preserving

Adversarial Robustness for Tabular Data through Cost and Utility Awareness

no code implementations27 Aug 2022 Klim Kireev, Bogdan Kulynych, Carmela Troncoso

We argue that, due to the differences between tabular data and images or text, existing threat models are not suitable for tabular domains.

Abuse Detection Adversarial Robustness

What You See is What You Get: Principled Deep Learning via Distributional Generalization

1 code implementation7 Apr 2022 Bogdan Kulynych, Yao-Yuan Yang, Yaodong Yu, Jarosław Błasiok, Preetum Nakkiran

In contrast, we show that Differentially-Private (DP) training provably ensures the high-level WYSIWYG property, which we quantify using a notion of distributional generalization.

Adversarial for Good? How the Adversarial ML Community's Values Impede Socially Beneficial Uses of Attacks

no code implementations ICML Workshop AML 2021 Kendra Albert, Maggie Delano, Bogdan Kulynych, Ram Shankar Siva Kumar

In this paper, we review the broader impact statements that adversarial ML researchers wrote as part of their NeurIPS 2020 papers and assess the assumptions that authors have about the goals of their work.

Exploring Data Pipelines through the Process Lens: a Reference Model forComputer Vision

no code implementations5 Jul 2021 Agathe Balayn, Bogdan Kulynych, Seda Guerses

Researchers have identified datasets used for training computer vision (CV) models as an important source of hazardous outcomes, and continue to examine popular CV datasets to expose their harms.

Questioning the assumptions behind fairness solutions

no code implementations27 Nov 2018 Rebekah Overdorf, Bogdan Kulynych, Ero Balsa, Carmela Troncoso, Seda Gürses

In addition to their benefits, optimization systems can have negative economic, moral, social, and political effects on populations as well as their environments.

Decision Making Fairness +1

Evading classifiers in discrete domains with provable optimality guarantees

2 code implementations25 Oct 2018 Bogdan Kulynych, Jamie Hayes, Nikita Samarin, Carmela Troncoso

We introduce a graphical framework that (1) generalizes existing attacks in discrete domains, (2) can accommodate complex cost functions beyond $p$-norms, including financial cost incurred when attacking a classifier, and (3) efficiently produces valid adversarial examples with guarantees of minimal adversarial cost.

Adversarial Robustness Spam detection +2

POTs: Protective Optimization Technologies

1 code implementation7 Jun 2018 Bogdan Kulynych, Rebekah Overdorf, Carmela Troncoso, Seda Gürses

Fairness frameworks do so, in part, by mapping these problems to a narrow definition and assuming the service providers can be trusted to deploy countermeasures.

Decision Making Fairness

Feature importance scores and lossless feature pruning using Banzhaf power indices

no code implementations14 Nov 2017 Bogdan Kulynych, Carmela Troncoso

In particular, we propose the use of the Banzhaf power index as a measure of influence of features on the outcome of a classifier.

Feature Importance General Classification

ClaimChain: Improving the Security and Privacy of In-band Key Distribution for Messaging

2 code implementations19 Jul 2017 Bogdan Kulynych, Wouter Lueks, Marios Isaakidis, George Danezis, Carmela Troncoso

Autocrypt is a new community-driven open specification for e-mail encryption that attempts to respond to this demand.

Cryptography and Security

Cannot find the paper you are looking for? You can Submit a new open access paper.