Bogdan Kulynych (@hiddenmarkov) - Twitter Profili

Presenting this on Thursday Dec 4 at #EurIPS in Copenhagen. Come by at the poster session if this sounds interesting! #NeurIPS2025

Bogdan Kulynych@hiddenmarkov

New paper at #NeurIPS2025! "Unifying Re-Identification, Attribute Inference, and Data Reconstruction Risks in Differential Privacy" in which we derive unified, tighter bounds on operational attack risks for any DP mechanisms, using f-DP. Link: arxiv.org/abs/2507.06969 Thread👇

English

0

1

255

Bogdan Kulynych@hiddenmarkov·1 Ara

Continued here: x.com/hiddenmarkov/s…

Bogdan Kulynych@hiddenmarkov

Continuing the thread on "Unifying Re-Identification, Attribute Inference, and Data Reconstruction Risks in Differential Privacy", for some reason it got borked. x.com/hiddenmarkov/s…

English

0

98

Bogdan Kulynych@hiddenmarkov·1 Ara

This is a unifying framework which can model various types of risk.

English

1

0

1

84

Bogdan Kulynych@hiddenmarkov·1 Ara

New paper at #NeurIPS2025! "Unifying Re-Identification, Attribute Inference, and Data Reconstruction Risks in Differential Privacy" in which we derive unified, tighter bounds on operational attack risks for any DP mechanisms, using f-DP. Link: arxiv.org/abs/2507.06969 Thread👇

English

2

8

675

Bogdan Kulynych@hiddenmarkov·1 Ara

Very excited, and I think this will be quite useful for practical deployments of DP. This is a joint work with great Felipe Gomez ( felipe-gomez.com ), George Kaissis, Jamie Hayes, Borja Balle, @FlavioCalmon, JL Raisaro.

English

0

54

Bogdan Kulynych@hiddenmarkov·1 Ara

Another (final) finding. The unified f-DP bound extends to a form of a generalization bound. Given that we can compute f-DP curves precisely, this is likely the tightest generalization bound applicable to deep learning, but it is only for on-average generalization unfortunately.

English

1

0

45

Bogdan Kulynych@hiddenmarkov·1 Ara

Continuing the thread on "Unifying Re-Identification, Attribute Inference, and Data Reconstruction Risks in Differential Privacy", for some reason it got borked. x.com/hiddenmarkov/s…

Bogdan Kulynych@hiddenmarkov

New paper at #NeurIPS2025! "Unifying Re-Identification, Attribute Inference, and Data Reconstruction Risks in Differential Privacy" in which we derive unified, tighter bounds on operational attack risks for any DP mechanisms, using f-DP. Link: arxiv.org/abs/2507.06969 Thread👇

English

1

0

1

206

Bogdan Kulynych

Keşfet