4 papers
Quantitative Gaussian-Process limits of Tensor Programs
Andrea Agazzi, Eloy Mosig García, Dario Trevisan
We study the infinite-width Gaussian-process limit of random neural networks through the lens of tensor programs, and we provide a quantitative convergence theory in Wasserstein di…
Beyond NNGP: Large Deviations and Feature Learning in Bayesian Neural Networks
Katerina Papagiannouli, Dario Trevisan, Giuseppe Pio Zitto
We study wide Bayesian neural networks focusing on the rare but statistically dominant fluctuations that govern posterior concentration, beyond Gaussian-process limits. Large-devia…
Quantitative convergence of trained single layer neural networks to Gaussian processes
Eloy Mosig, Andrea Agazzi, Dario Trevisan
In this paper, we study the quantitative convergence of shallow neural networks trained via gradient descent to their associated Gaussian processes in the infinite-width limit. Whi…
Wide Deep Neural Networks with Gaussian Weights are Very Close to Gaussian Processes
Dario Trevisan
We establish novel rates for the Gaussian approximation of random deep neural networks with Gaussian parameters (weights and biases) and Lipschitz activation functions, in the wide…