Differential privacy | The Alan Turing Institute

Projects

Person walking up stairs

Ground truth for mental health data science

Linking digital footprint data in the UK birth cohorts to guide the next generation of algorithms for mental health and wellbeing research

banknotes

Synthetic data generation for finance and economics

Generating artificial business, finance, and economic data to enable data sharing, scenario simulation, and model validation

Coloured panels

Trustworthy digital infrastructure for identity systems

Enhancing the privacy and security of national digital identity systems

Blurred people walking

QUIPP – Quantifying utility and preserving privacy in synthetic data sets

Understanding the balance between utility, privacy and the uncertainty associated with synthetic data sets

Coffee and receipts

Donating personal transactional data for research

Investigating the public acceptability of using commercial transactional data in public health research

Storage lockups

Statistical secret sharing

Developing methodology for multiple parties to securely share and unify their statistical inference, without having to reveal their inference or underlying data

Windows

Evaluating privacy-preserving generative models in the wild

Evaluating existing work on privacy-preserving synthetic data generation

Programmes

Enigma machine

Defence and security

Collaborating with the defence and security community to deliver an ambitious programme of data science and artificial intelligence research

Publications and software

SecureABC: Secure antibody certificates for COVID-19

COVID-19 has resulted in unprecedented social distancing policies being enforced worldwide. As governments urgently...

Chris Hicks*, David Butler*, Carsten Maple, Jon Crowcroft. SecureABC: Secure AntiBody Certificates for COVID-19. CoRR, abs/2005.11833. 2020.

Differentially private health tokens for estimating COVID-19 risk

In the fight against Covid-19, many governments and businesses are in the process of...

David Butler*, Chris Hicks*, James Bell, Carsten Maple, Jon Crowcroft. Differentially Private Health Tokens for Estimating COVID-19 Risk. appeared in NDSS CoronaDef 2021.

Events

group work with laptops

CIFAR-UKRI-CNRS AI & society: From principles to practice

Monday 03 Jun 2019 - Tuesday 04 Jun 2019

Time: 09:00 - 18:30

Interest groups

abstract background

Synthetic data

How can synthetic data be used to correct bias in data?

radio waves

Machine learning for radio frequency applications

How do we generate robust, real-time machine learning algorithms with integrated hardware and software for radio frequency applications?

Connecting lines and dots

AI in resource- and data-constrained environments

How can we build effective predictive models on hardware platforms that have limited computational resources and/or when the model training data are by definition always limited in amount?

A busy crowd of people with a blurred effect

Trustworthy digital identity

How can we understand, evaluate and advance the trustworthiness of digital identity systems?

Pins linked with string

Analysis of networks

What problems of mutual interest are there for researchers interested in network analysis?

Locked bolt on wooden door

Privacy-preserving data analysis

How can we enable personal data to be shared without compromising our privacy?

People

badat

Rafiah Patel

Enrichment Student

teddy

Teddy Cunningham

Enrichment Student

Samuel Kaski

Professor Samuel Kaski

Turing Fellow

Maxine Mackintosh

Maxine Mackintosh

Programme Lead – Diverse Data, Genomics England

Professor Andrea Cavallaro

Professor Andrea Cavallaro

Turing Fellow

Andrew Dowsey

Professor Andrew Dowsey

Turing Fellow

Anya Skatova

Dr Anya Skatova

Turing Fellow

Murray Pollock

Dr Murray Pollock

Turing Fellow

Sylvie Delacroix

Professor Sylvie Delacroix

Turing Fellow