Brightlands Techruption MPC Day

Event date: December 04, 201909:30u
End date: December 04, 2019


You are invited to join the Brightlands Techruption MPC day on December 4th, 2019 at the Brightlands Smart Services Campus in Heerlen. The goal of the day is to jointly evaluate the results of the running use cases in the MPC&L (Multi-Party Computation & Learning) track: Combatting Poverty, Confidential Benchmarking and Fraud Detection.  Multi-Party Computation, and related techniques like federated learning, enable organizations to jointly analyze (and learn from) their sensitive data without having to share or reveal this data. We will show and explore how these techniques allow parties to collaborate without sharing privacy-sensitive data. On top of this we would like to gather your thoughts and ideas on possible continuations of these use cases, as well as brainstorming on new ideas to be started in the MPC&L track.

09.30    Coffee and short introduction into MPC&L
10.00    Stagegate Combatting Poverty (this stagegate will be in Dutch)
11.00    Stagegate Fraud Detection
12.00    Lunch
13.00    Stagegate Confidential Benchmarking
14.00    Ideation new MPC&L use case(s) 
                 •    Do you have a problem that could be solved using another party’s data?
                 •    Multi-Party Computation & Learning may offer opportunities to tackle such a problem in a
                      privacy-preserving manner.
15.00    Decisions and action points
15.30    Drinks

On this day we will report on:

  • Combatting poverty:

Poverty is a challenging problem in many municipalities. There is a strong motivation to detect poverty early, or even prevent it.  This is relevant for the municipality of Heerlen, which often ranks high on poverty lists, but also a broader societal challenge. Generally the approach to tackle poverty is after-the-fact: it starts when someone is already in poverty. There is too little insight into the influence of other factors on poverty and payment problems, such as the use of utilities, health costs, income and demographic factors, because the underlying data is privacy-sensitive and cannot be shared. This use-case explores the extent to which the municipal government could use Multi-Party Computation to gain relevant insights into the different factors influencing poverty without revealing or sharing data at the individual level, preserving privacy of individuals. 
This use case is interesting for organizations that wish to contribute to combatting poverty and have (potentially) relevant data that may be related to poverty issues.

  • Confidential benchmarking:

For many application domains benchmarking is relevant, specifically to compare the position of an organisation (e.g. a health institute, bank, freelancer) with that of competing organisations. Using trusted third parties is often slow, expensive and in some cases not feasible due to the confidentiality or privacy-sensitivity of the data. MPC and related techniques could play an important role for confidential benchmarking applications, removing the need for a trusted third party. Different applications have been explored in this use-case, including confidential benchmarking in health, and confidential benchmarking of diversity in companies. Furthermore, first steps have been taken to investigate feasibility for a generic confidential benchmarking tool using MPC and related techniques.
This use case is interesting for organizations with interest in benchmarking in health or w.r.t. diversity, but also for organizations with other needs for benchmarking applications.

  • Fraud detection:

Often, insurance fraud by healthcare providers is discovered (too) late by health insurance companies, because they lack crucial information known to other parties like banks and the Chamber of Commerce (KvK). Combining fraud indicators may give sufficient cause to suspect and investigate fraud, but this data cannot be readily shared between these parties due to confidentiality of the data. MPC offers a solution to securely combine the fraud indicators, without revealing the actual confidential indicators. This allows the participating parties to identify which companies need to be investigated further. A Proof-of-Concept has been developed to demonstrate the MPC solution. Also, this solution has potential for detection of other forms of fraud where information needs to be combined.
This use case is interesting for organizations with interest in using MPC to combine fraud indicators with other organizations in a secure way.

On this day, we’ll also collect and discuss ideas for new MPC&L use cases, so your ideas are very welcome. A specific question to consider: do you have a problem that could be solved using another party’s data?
The Techruption MPC&L Day is open for all Techruption parties active or interested in the Techruption program, and specifically MPC and related techniques like federated learning.
Please register so we can plan accordingly the catering, room reservation, announce your visit at the front desk etc.

  overview news & events