collaborators

6 papers

cs.LG2024

Of Dice and Games: A Theory of Generalized Boosting

Marco Bressan, Nataly Brukhim, Nicolò Cesa-Bianchi +4

Cost-sensitive loss functions are crucial in many real-world prediction problems, where different types of errors are penalized differently; for example, in medical diagnosis, a fa…

cs.LG2024

Improved Regret Bounds for Bandits with Expert Advice

Nicolò Cesa-Bianchi, Khaled Eldowa, Emmanuel Esposito +1

In this research note, we revisit the bandits with expert advice problem. Under a restricted feedback model, we prove a lower bound of order for the worst-cas…

cs.LG2024

Efficient Algorithms for Learning Monophonic Halfspaces in Graphs

Marco Bressan, Emmanuel Esposito, Maximilian Thiessen

We study the problem of learning a binary classifier on the vertices of a graph. In particular, we consider classifiers given by monophonic halfspaces, partitions of the vertices t…

cs.LG2024

A Theory of Interpretable Approximations

Marco Bressan, Nicolò Cesa-Bianchi, Emmanuel Esposito +3

Can a deep neural network be approximated by a small decision tree based on simple features? This question and its variants are behind the growing demand for machine learning model…

cs.LG2023

An Improved Uniform Convergence Bound with Fat-Shattering Dimension

Roberto Colomboni, Emmanuel Esposito, Andrea Paudice

The fat-shattering dimension characterizes the uniform convergence property of real-valued functions. The state-of-the-art upper bounds feature a multiplicative squared logarithmic…

cs.LG2023

Delayed Bandits: When Do Intermediate Observations Help?

Emmanuel Esposito, Saeed Masoudian, Hao Qiu +3

We study a -armed bandit with delayed feedback and intermediate observations. We consider a model where intermediate observations have a form of a finite state, which is observe…