59 citations · 195 across the 37 of their papers we have counts for
40 papers
Learning Ergodic Dynamical Systems from a Finite Trajectory
Oleksii Kachaiev, Silvia Villa, Lorenzo Rosasco
We consider the problem of learning from a single finite trajectory of an ergodic stochastic dynamical system. More precisely, we study discrete-time autonomous stochastic systems…
Optimization Insights into Deep Diagonal Linear Networks
Hippolyte Labarrière, Cesare Molinari, Lorenzo Rosasco +2
Gradient-based methods successfully train highly overparameterized models in practice, even though the associated optimization problems are markedly nonconvex. Understanding the me…
On Learning the Optimal Regularization Parameter in Inverse Problems
Jonathan Chirinos Rodriguez, Ernesto De Vito, Cesare Molinari +2
Selecting the best regularization parameter in inverse problems is a classical and yet challenging problem. Recently, data-driven approaches have become popular to tackle this chal…
Variance reduction techniques for stochastic proximal point algorithms
Cheik Traoré, Vassilis Apidopoulos, Saverio Salzo +1
In the context of finite sums minimization, variance reduction techniques are widely used to improve the performance of state-of-the-art stochastic gradient methods. Their practica…
Regularization properties of dual subgradient flow
Vassilis Apidopoulos, Cesare Molinari, Lorenzo Rosasco +1
Dual gradient descent combined with early stopping represents an efficient alternative to the Tikhonov variational approach when the regularizer is strongly convex. However, for ma…
An Optimal Structured Zeroth-order Algorithm for Non-smooth Optimization
Marco Rando, Cesare Molinari, Lorenzo Rosasco +1
Finite-difference methods are a class of algorithms designed to solve black-box optimization problems by approximating a gradient of the target function on a set of directions. In…