Showing 2026Show all
3 papers · 1 filter
math.PR2026
Pathological Large Deviations of the KMP Process in Dimension
Daniel Heydecker
We study dynamic large deviations for the Kipnis--Marchioro--Presutti process on the discrete torus . By recasting the candidate rate function in terms of a linear…
cs.LG2026
Large Spikes in Stochastic Gradient Descent: A Large-Deviations View
Benjamin Gess, Daniel Heydecker
Large loss spikes in stochastic gradient descent are studied through a rigorous large-deviations analysis for a shallow, fully connected network in the NTK scaling. In contrast to…
math.PR2026
The Porous Medium Equation: Multiscale Integrability in Large Deviations
Benjamin Gess, Daniel Heydecker
We consider a zero-range process with superlinear local jump rate, which in a hydrodynamic-small particle rescaling converges to the porous medium equation $\partial_t…