Personal information

Emails & domains

Verified email domains

Websites & social links

Personal Website

Countries

Germany

Activities

Collapse all

Education and qualifications (3)

Technical University of Munich: Munich, DE

2020-10-12 to 2024 | PhD Student (AI in Medicine)
Education
Source: Self-asserted source
Dmitrii Usynin

Imperial College London: London, London, GB

2020-10-12 to 2024 | PhD Student (Department of Computing)
Education
Source: Self-asserted source
Dmitrii Usynin

Imperial College London: London, GB

2016-10-01 to 2020-06-02 | MEng Computing (Department of Computing)
Education
Source: Self-asserted source
Dmitrii Usynin

Works (15)

Incentivising the federation: gradient-based metrics for data selection and valuation in private decentralised training

2024-06-05 | Conference paper
DOI: 10.1145/3655693.3660253
Contributors: Dmitrii Usynin; Daniel Rueckert; Georgios Kaissis
Source: check_circle
Crossref

Differentially Private Guarantees for Analytics and Machine Learning on Graphs: A Survey of Results

Journal of Privacy and Confidentiality
2024-02-11 | Journal article
DOI: 10.29012/jpc.820
Contributors: Tamara T. Mueller; Dmitrii Usynin; Johannes C. Paetzold; Rickmer Braren; Daniel Rueckert; Georgios Kaissis
Source: check_circle
Crossref

Membership Inference Attacks Against Semantic Segmentation Models

2023-11-30 | Conference paper
DOI: 10.1145/3605764.3623906
Contributors: Tomas Chobola; Dmitrii Usynin; Georgios Kaissis
Source: check_circle
Crossref

Beyond Gradients: Exploiting Adversarial Priors in Model Inversion Attacks

ACM Transactions on Privacy and Security
2023-08-30 | Journal article
DOI: 10.1145/3592800
Contributors: Dmitrii Usynin; Daniel Rueckert; Georgios Kaissis
Source: check_circle
Crossref

Unified Interpretation of the Gaussian Mechanism for Differential Privacy Through the Sensitivity Index

Journal of Privacy and Confidentiality
2022-07-29 | Journal article
DOI: 10.29012/jpc.807
Contributors: Georgios Kaissis; Moritz Knolle; Friederike Jungmann; Alexander Ziller; Dmitrii Usynin; Daniel Rueckert
Source: check_circle
Crossref

Medical imaging deep learning with differential privacy

Scientific Reports
2021-12 | Journal article
DOI: 10.1038/s41598-021-93030-0
Part of ISSN: 2045-2322
Source: Self-asserted source
Dmitrii Usynin

Adversarial interference and its mitigations in privacy-preserving collaborative machine learning

Nature Machine Intelligence
2021-09 | Journal article
DOI: 10.1038/s42256-021-00390-3
Part of ISSN: 2522-5839
Source: Self-asserted source
Dmitrii Usynin
grade
Preferred source (of 2)‎

A unified interpretation of the Gaussian mechanism for differential privacy through the sensitivity index

2021-09-22 | Preprint
ARXIV: arXiv:2109.10528v1
Source: Self-asserted source
Dmitrii Usynin

An automatic differentiation system for the age of differential privacy

2021-09-22 | Preprint
ARXIV: arXiv:2109.10573v1
Source: Self-asserted source
Dmitrii Usynin

Partial sensitivity analysis in differential privacy

2021-09-22 | Preprint
ARXIV: arXiv:2109.10582v1
Source: Self-asserted source
Dmitrii Usynin

Differentially private training of neural networks with Langevin dynamics for calibrated predictive uncertainty

2021-07-09 | Preprint
ARXIV: arXiv:2107.04296v1
Source: Self-asserted source
Dmitrii Usynin

Sensitivity analysis in differentially private machine learning using hybrid automatic differentiation

2021-07-09 | Preprint
ARXIV: arXiv:2107.04265v1
Source: Self-asserted source
Dmitrii Usynin

Differentially private federated deep learning for multi-site medical image segmentation

2021-07-06 | Preprint
ARXIV: arXiv:2107.02586v1
Source: Self-asserted source
Dmitrii Usynin

End-to-end privacy preserving deep learning on multi-institutional medical imaging

Nature Machine Intelligence
2021-06 | Journal article
DOI: 10.1038/s42256-021-00337-8
Part of ISSN: 2522-5839
Source: Self-asserted source
Dmitrii Usynin
grade
Preferred source (of 2)‎

Privacy-preserving medical image analysis

2020-12-10 | Preprint
ARXIV: arXiv:2012.06354v1
Source: Self-asserted source
Dmitrii Usynin
How many people are using ORCID?