2 papers
math.OC2025
Convergence of Markov Chains for Constant Step-size Stochastic Gradient Descent with Separable Functions
David Shirokoff, Philip Zaleski
Stochastic gradient descent (SGD) is a popular algorithm for minimizing objective functions that arise in machine learning. For constant step-sized SGD, the iterates form a Markov…
math.AP2024
A variational model of charged drops in dielectrically matched binary fluids: the effect of charge discreteness
Cyrill B. Muratov, Matteo Novaga, Philip Zaleski
This paper addresses the ill-posedness of the classical Rayleigh variational model of conducting charged liquid drops by incorporating the discreteness of the elementary charges. I…