Search Results for author: Luca Bortolussi

Found 26 papers, 9 papers with code

Conditioning Score-Based Generative Models by Neuro-Symbolic Constraints

no code implementations31 Aug 2023 Davide Scassola, Sebastiano Saccani, Ginevra Carbone, Luca Bortolussi

Score-based and diffusion models have emerged as effective approaches for both conditional and unconditional generation.

Time Series

CARSO: Blending Adversarial Training and Purification Improves Adversarial Robustness

1 code implementation25 May 2023 Emanuele Ballarin, Alessio Ansuini, Luca Bortolussi

In this work, we propose a novel adversarial defence mechanism for image classification - CARSO - blending the paradigms of adversarial training and adversarial purification in a mutually-beneficial, robustness-enhancing way.

Adversarial Defense Adversarial Robustness +2

Relating Implicit Bias and Adversarial Attacks through Intrinsic Dimension

1 code implementation24 May 2023 Lorenzo Basile, Nikos Karantzas, Alberto D'Onofrio, Luca Bortolussi, Alex Rodriguez, Fabio Anselmi

Despite their impressive performance in classification, neural networks are known to be vulnerable to adversarial attacks.

Image Classification

Towards Invertible Semantic-Preserving Embeddings of Logical Formulae

no code implementations3 May 2023 Gaia Saveri, Luca Bortolussi

State of the art Machine Learning (ML) approaches are mostly based on gradient descent optimisation in continuous spaces, while learning logic is framed in the discrete syntactic space of formulae.

Conformal Quantitative Predictive Monitoring of STL Requirements for Stochastic Processes

1 code implementation4 Nov 2022 Francesca Cairoli, Nicola Paoletti, Luca Bortolussi

We consider the problem of predictive monitoring (PM), i. e., predicting at runtime the satisfaction of a desired property from the current system's state.

Prediction Intervals

On the Robustness of Bayesian Neural Networks to Adversarial Attacks

1 code implementation13 Jul 2022 Luca Bortolussi, Ginevra Carbone, Luca Laurenti, Andrea Patane, Guido Sanguinetti, Matthew Wicker

Despite significant efforts, both practical and theoretical, training deep learning models robust to adversarial attacks is still an open problem.

Variational Inference

Scalable Stochastic Parametric Verification with Stochastic Variational Smoothed Model Checking

no code implementations11 May 2022 Luca Bortolussi, Francesca Cairoli, Ginevra Carbone, Paolo Pulcini

As observations are costly and noisy, smMC is framed as a Bayesian inference problem so that the estimates have an additional quantification of the uncertainty.

Bayesian Inference Gaussian Processes +1

Graph Neural Networks for Propositional Model Counting

no code implementations9 May 2022 Gaia Saveri, Luca Bortolussi

Graph Neural Networks (GNNs) have been recently leveraged to solve several logical reasoning tasks.

Logical Reasoning

Neural Predictive Monitoring under Partial Observability

1 code implementation16 Aug 2021 Francesca Cairoli, Luca Bortolussi, Nicola Paoletti

We consider the problem of predictive monitoring (PM), i. e., predicting at runtime future violations of a system from the current state.

Active Learning Conformal Prediction

Abstraction of Markov Population Dynamics via Generative Adversarial Nets

1 code implementation24 Jun 2021 Francesca Cairoli, Ginevra Carbone, Luca Bortolussi

Markov Population Models are a widespread formalism used to model the dynamics of complex systems, with applications in Systems Biology and many other fields.

Abstraction-Guided Truncations for Stationary Distributions of Markov Population Models

no code implementations3 May 2021 Michael Backenköhler, Luca Bortolussi, Gerrit Großmann, Verena Wolf

To understand the long-run behavior of Markov population models, the computation of the stationary distribution is often a crucial part.

Resilience of Bayesian Layer-Wise Explanations under Adversarial Attacks

1 code implementation22 Feb 2021 Ginevra Carbone, Guido Sanguinetti, Luca Bortolussi

We empirically show that interpretations provided by Bayesian Neural Networks are considerably more stable under adversarial perturbations of the inputs and even under direct attacks to the explanations.

General Classification

Random Projections for Improved Adversarial Robustness

no code implementations18 Feb 2021 Ginevra Carbone, Guido Sanguinetti, Luca Bortolussi

We propose two training techniques for improving the robustness of Neural Networks to adversarial attacks, i. e. manipulations of the inputs that are maliciously crafted to fool networks into incorrect predictions.

Adversarial Robustness Dimensionality Reduction

Analysis of Markov Jump Processes under Terminal Constraints

1 code implementation20 Oct 2020 Michael Backenköhler, Luca Bortolussi, Gerrit Großmann, Verena Wolf

Many probabilistic inference problems such as stochastic filtering or the computation of rare event probabilities require model analysis under initial and terminal constraints.

Bayesian Inference

Adversarial Learning of Robust and Safe Controllers for Cyber-Physical Systems

no code implementations4 Sep 2020 Luca Bortolussi, Francesca Cairoli, Ginevra Carbone, Francesco Franchina, Enrico Regolin

We introduce a novel learning-based approach to synthesize safe and robust controllers for autonomous Cyber-Physical Systems and, at the same time, to generate challenging tests.

Density-embedding layers: a general framework for adaptive receptive fields

no code implementations23 Jun 2020 Francesco Cicala, Luca Bortolussi

In this paper, we propose the novel theoretical framework of density-embedded layers, generalizing the transformation represented by a neuron.

Improved estimations of stochastic chemical kinetics by finite state expansion

no code implementations12 Jun 2020 Tabea Waizmann, Luca Bortolussi, Andrea Vandin, Mirco Tribastone

The deterministic rate equation (DRE) gives a macroscopic approximation as a compact system of differential equations that estimate the average populations for each species, but it may be inaccurate in the case of nonlinear interaction dynamics.

Robustness of Bayesian Neural Networks to Gradient-Based Attacks

1 code implementation NeurIPS 2020 Ginevra Carbone, Matthew Wicker, Luca Laurenti, Andrea Patane, Luca Bortolussi, Guido Sanguinetti

Vulnerability to adversarial attacks is one of the principal hurdles to the adoption of deep learning in safety-critical applications.

Variational Inference

Intrinsic Geometric Vulnerability of High-Dimensional Artificial Intelligence

no code implementations8 Nov 2018 Luca Bortolussi, Guido Sanguinetti

The success of modern Artificial Intelligence (AI) technologies depends critically on the ability to learn non-linear functional dependencies from large, high dimensional data sets.

Vocal Bursts Intensity Prediction

A Robust Genetic Algorithm for Learning Temporal Specifications from Data

no code implementations13 Nov 2017 Laura Nenzi, Simone Silvetti, Ezio Bartocci, Luca Bortolussi

We consider the problem of mining signal temporal logical requirements from a dataset of regular (good) and anomalous (bad) trajectories of a dynamical system.

Matching models across abstraction levels with Gaussian Processes

no code implementations7 May 2016 Giulio Caravagna, Luca Bortolussi, Guido Sanguinetti

Biological systems are often modelled at different levels of abstraction depending on the particular aims/resources of a study.

Gaussian Processes

Learning Temporal Logical Properties Discriminating ECG models of Cardiac Arrhytmias

no code implementations29 Dec 2013 Ezio Bartocci, Luca Bortolussi, Guido Sanguinetti

We present a novel approach to learn the formulae characterising the emergent behaviour of a dynamical system from system observations.

On the Robustness of Temporal Properties for Stochastic Models

no code implementations3 Sep 2013 Ezio Bartocci, Luca Bortolussi, Laura Nenzi, Guido Sanguinetti

By discussing two examples, we show how to approximate the distribution of the robustness score and its key indicators: the average robustness and the conditional average robustness.

Cannot find the paper you are looking for? You can Submit a new open access paper.