Skip to content

Alessio Benavoli

MainMenu MainMenu
  • Home
  • Research
  • Publications
  • Software
  • Contacts
Expand Search Form

Software

2021-01-28 21:22:51

19

Skew Gaussian Processes

Skew Gaussian Processes

Skew Gaussian Processes allow a closed-form expression for the posterior and predictive distribution for regression, classification, preference and mixed problems with computation complexity of O(n^3) and storage demands of O(n^2) (same as for GP regression). This software library provides a unified framework for closed-form nonparametric inference for a large class of supervised learning problems.

View More

2018-03-15 13:30:30

17

BayComp (official Python library)

BayComp (official Python library)

BayComp is a library for Bayesian hypothesis testing for machine learning. This project aims to use modern algorithms ("Dirichlet process", "Markov chain Monte Carlo") to apply Bayesian analysis for assessing/comparing algorithms performance in machine learning. We have implemented the tests in Python and they can be installed via pip. We use Ipython notebooks to illustrate SW use.

View More

2017-06-10 02:23:07

15

Bayesian tests for machine learning

Bayesian tests for machine learning

This project aims to use modern algorithms ("Dirichlet process", "Markov chain Monte Carlo") to apply Bayesian analysis for assessing/comparing algorithms performance in machine learning. We have implemented the tests in R, Python, Julia and STAN. We use Ipython notebooks to illustrate SW use.

View More

2020-02-21 21:38:28

18

Uncertainty modeling for machine reasoning

Uncertainty modeling for machine reasoning

Matlab code and functions for the testing scenarios analysed in "A tutorial on uncertainty modeling for machine reasoning" The tutorial reviews the prevalent methods for model-based autonomous decision making based on observations and prior knowledge, primarily in the context of classification. The paper covers the main approaches to uncertain knowledge representation and reasoning, in particular: Bayesian probability theory, Possibility theory, Dempster-Shafer belief functions theory (using transferrable belief model) and Imprecise probability theory

View More

2017-06-10 02:23:07

16

The Imprecise Dirichlet Process Statistical Package

The Imprecise Dirichlet Process Statistical Package

Bayesian methods are ubiquitous in many research areas. Nevertheless, the analysis of empirical results is typically performed by frequentist tests. This implies dealing with null hypothesis significance tests and p-values, even though the shortcomings of such methods are well known. The IDP Statistical Package aims to change this perspective by providing Bayesian nonparametric versions of the most used frequentist tests. IDP is based on a Prior Near-Ignorance Dirichlet Process.

View More

PhD position

I have a fully-funded PhD position (part of the SFI-CRT in AI) on the development of machine learning methods for large scale time series extrinsic regression (more details here). Reach out for me if you want to chat about doing a PhD at Trinity College Dublin! The next deadline to submit an application is February 29th 2023.

Highlighted publications

Alessio Benavoli, Dario Azzimonti, Dario Piga: A unified framework for closed-form nonparametric regression, classification, preference and mixed problems with Skew Gaussian Processes. In: Machine Learning, pp. 1-39, 2021.
Benavoli , Alessio, Facchini, Alessandro, Zaffalon, Marco: The Weirdness Theorem and the Origin of Quantum Paradoxes. In: Foundations of Physics, vol. 51, no. 95, 2021.
Benavoli, Alessio, Corani, Giorgio, Demsar, Janez, Zaffalon, Marco: Time for a change: a tutorial for comparing multiple classifiers through Bayesian analysis. In: Journal of Machine Learning Research, vol. 18, no. 77, pp. 1-36, 2017.

Links

SCSS

Alessio Benavoli © 2023