Bienvenue sur mon site web personnel.

Voir mon cv

Papers

Conferences

• On the Alignment of Group Fairness with Attribute Privacy
• Privacy-Preserving Pseudonyms for LoRaWAN
• MixNN: Protection of Federated Learning Against Inference Attacks by Mixing Neural Network Layers

Manuscrit de thèse

Programmation

You can browse through all my repos here: Browse my git repos.

Selection of projects

Finit classification

License GPLv3

Efficient classification algorithm on finite sets. We use combinatorics to find an application between two finite sets such that the balanced accuracy is the highest among all possible applications.

• Read the documentation
• Browse the source code

Probablity law of the number of coin flips to get m consecutive heads

License GPLv3

We compute the exact probability mass function of the number of coin flips to get $m$ consecutive heads. Meaning that we take D, a random variable modeling the number of coin flips until the coin gives $m$ heads in a row; then we explicit a function f such that for all n, f(n) = P(D=n). P(D=n) is the probability that the m last flips among n total flips are heads while the n-m first flips doesn't contain a string of m uninterrupted heads. Using f we compute the cumulative distribution function (CDF) of D as well as E(D): the expected number of flips.

• Read the documentation
• Browse the source code