Skip to main content

New Paper: Double Descent in Particle Physics Data

Author
Lukas Heinrich
Lab Lead

We’re happy to announce a new paper from our research lab. The paper is the first to discover and invesitgate the “Double Descent Phenomenon” in particle physics data. Check it out! https://arxiv.org/abs/2509.01397

Abstract
#

Recently, the benefit of heavily overparameterized models has been observed in machine learning tasks: models with enough capacity to easily cross the interpolation threshold improve in generalization error compared to the classical bias-variance tradeoff regime. We demonstrate this behavior for the first time in particle physics data and explore when and where ‘double descent’ appears and under which circumstances overparameterization results in a performance gain.