publications

Concentration inequalities under sub-Gaussian and sub-exponential conditions

with Massimiliano Pontil,

Robust unsupervised learning via L-statistic minimization

with Daniela Angela Parletta, Andrea Paudice, Massimiliano Pontil,

Estimating weighted areas under the ROC curve

with Massimiliano Pontil,

Exploiting MMD and Sinkhorn Divergences for Fair and Transferable Representation Learning

with Luca Oneto, Michele Donini, Giulia Luise, Carlo Ciliberto, Massimiliano Pontil,

Uniform concentration and symmetrization for weak interactions

with Massimiliano Pontil,

Empirical bounds for functions with weak interactions

with Massimiliano Pontil,

A second-order look at stability and generalization

A Bernstein-type inequality for functions of bounded interaction

A vector-contraction inequality for Rademacher complexities

The benefit of multitask representation learning

with Massimiliano Pontil and Bernardino Romera-Paredes

A chain rule for the expected suprema of Gaussian processes

An inequality with applications to structured sparsity and multitask dictionary learning

with Massimiliano Pontil and Bernardino Romera-Paredes

Excess risk bounds for multitask learning with trace norm regularization

with Massimiliano Pontil

Sparse coding for multitask and transfer learning

with Massimiliano Pontil and Bernardino Romera-Paredes

Structured sparsity and generalization

with Massimiliano Pontil

Thermodynamics and Concentration

K-dimensional coding schemes in Hilbert spaces

long vers. with M.Pontil

Dominated Concentration

Empirical Bernstein Bounds and Sample Variance Penalization

with Massimiliano Pontil

Transfer bounds for linear feature learning

Unsupervised Slow Subspace-Learning from Stationary Processes

long version

Learning Similarity with Operator-valued Large-margin Classifiers

A uniform lower error bound for half-space learning

with Massimiliano Pontil

Uniform error bounds for K-dimensional coding schemes in Hilbert spaces

with Massimiliano Pontil

An Algorithm for Transfer Learning in a Heterogeneous Environment

with A.Argyriou and M.Pontil

Learning to Compare using Operator-Valued Large-Margin Classifiers

The Rademacher Complexity of Linear Transformation Classes

Unsupervised Slow Subspace-Learning from Stationary Processes

Bounds for Linear Multi-Task Learning

Generalization Bounds for Subspace Selection and Hyperbolic PCA

Algorithmic Stability and Meta-Learning

Concentration Inequalities for Functions of Independent Variables

A Bound on the Deviation Probability for Sums of Non-Negative Random Variables

Phase transitions and positive definite couplings

back to my homepage