4 papers · 1 filter
Functional Large Deviations for Wide Deep Neural Networks with Gaussian Initialization and Lipschitz Activations
Claudio Macci, Barbara Pacchiarotti, Katerina Papagiannouli +2
We establish a functional large deviation principle for fully connected multi-layer perceptrons with i.i.d. Gaussian weights (LeCun initialization) and general Lipschitz activation…
Asymptotics for Random Quadratic Transportation Costs
Martin Huesmann, Michael Goldman, Dario Trevisan
We establish the validity of asymptotic limits for the general transportation problem between random i.i.d. points and their common distribution, with respect to the squared Euclid…
Wasserstein asymptotics for empirical measures of diffusions on four dimensional closed manifolds
Dario Trevisan, Feng-Yu Wang, Jie-Xiang Zhu
We identify the leading term in the asymptotics of the quadratic Wasserstein distance between the invariant measure and empirical measures for diffusion processes on closed weighte…
Subadditivity and optimal matching of unbounded samples
Emanuele Caglioti, Michael Goldman, Francesca Pieroni +1
We obtain new bounds for the optimal matching cost for empirical measures with unbounded support. For a large class of radially symmetric and rapidly decaying probability laws, we…