Showing math.PRShow all
2 papers · 1 filter
math.PR2024
Large deviations of one-hidden-layer neural networks
Christian Hirsch, Daniel Willhalm
We study large deviations in the context of stochastic gradient descent for one-hidden-layer neural networks with quadratic loss. We derive a quenched large deviation principle, wh…
math.PR2023
Lower large deviations for geometric functionals in sparse, critical and dense regimes
Christian Hirsch, Daniel Willhalm
We prove lower large deviations for geometric functionals in sparse, critical and dense regimes. Our results are tailored for functionals with nonexisting exponential moments, for…