Decentralised Finance (DeFi) refers to applications running on public blockchains that aim to replicate and extend some traditional financial functionalities without relying on centralised intermediaries. Over the last decade, DeFi has grown at a rapid pace, giving rise to algorithmic and financial innovations with spillover effects beyond the boundaries of crypto markets alone.

Unlike traditional finance (TradFi), which depends on banks, brokers, and regulated exchanges, DeFi protocols operate through smart contracts: self-executing code deployed on a blockchain. This enables permissionless lending and borrowing, accessible staking, and has led to the emergence of decentralised exchanges taking unprecedented forms — all running transparently and continuously, without human intermediaries. While this opens remarkable possibilities, it also introduces new theoretical and practical questions.

The mathematical and statistical modelling of DeFi protocols remains in its infancy. Rigorous quantitative methods are needed to understand the structural properties of these systems, assess their risks, and design more robust next-generation protocols. The ANR BlockFi project addresses this gap by fostering the development of mathematical models, statistical tools, and formal verification methods. Topics under study include the design of interest rate models on lending protocols, the mechanics of automated market makers (AMMs), the application of formal proofs to protocols, asset management questions in the DeFi world, and the study of systemic risk.

To tackle these challenges, we have assembled an interdisciplinary team of mathematicians, statisticians, economists, and computer scientists. Beyond producing research, our ambition is to make these emerging scientific questions visible to the broader academic community and to train the next generation of researchers.

Project start September 2024
Duration 60 months
ANR reference ANR-24-CE38-7885
Charles Bertucci

Charles Bertucci

CNRS Researcher in Mathematics at Université Paris Dauphine-PSL. Expert in Partial Differential Equations, Modelling and Game Theory.

Louis Bertucci

Louis Bertucci

Research Fellow at Institut Louis Bachelier. Expert in Blockchain and DeFi.

Stefano De Marco

Stefano De Marco

Professor of Applied Mathematics at Ecole Polytechnique. Expert in Stochastic Modelling and Financial Risk Management.

Mnacho Echenim

Mnacho Echenim

Professor of Computer Science at Grenoble INP - Ensimag. Expert in Automated Theorem Proving and its Applications.

Emmanuel Gobet

Emmanuel Gobet

Professor of Probability and Statistics at Sorbonne Université. Expert in Risk Modelling and Quantitative Finance/Insurance.

Pierre-Olivier Goffard

Pierre-Olivier Goffard

Associate Professor of Applied Mathematics and Actuarial Science at Université de Strasbourg. Expert in Risk Modelling and Blockchain.

Olivier Guéant

Olivier Guéant

Professor of Applied Mathematics, Université Paris Cité. Expert in Applied Probability and Quantitative Finance.

Nicolas Peltier

Nicolas Peltier

CNRS Researcher in Computer Science. Expert in Automated Reasoning and Logic.

Julien Prat

Julien Prat

Director of Research in Economics, CNRS and IP Paris. Expert in Quantitative Economics.

January 7, 2025

BLOCKFI Launch Day

Official kick-off of the ANR BLOCKFI at Palais Brongniart (Paris).

Slides

May 13, 2025

DeFi Workshop

Workshop organised by the ANR BLOCKFI at Maison des Sciences Economiques (Paris).

Slides

December 15–16, 2025

DeFi Workshop

Two-day workshop organised by the ANR BLOCKFI at Laboratoire d'Informatique de Grenoble.

Slides

April 3, 10 & 17, 2025

DeFi Minicourse

Three sessions of two talks on decentralized finance at Institut Henri Poincaré (Paris).

Preprints

M. Echenim, E. Gobet, A.C. Maurice, Thorough mathematical modeling and analysis of Uniswap v3. In revision for SIAM Journal on Financial Mathematics, 2025. HAL
P.O. Goffard & Stéphane Loisel, Collaborative and parametric insurance on the Ethereum blockchain, 2024. HAL

Published

Coming soon...