18 citations · 27 across the 14 of their papers we have counts for
14 papers
From Atoms to Entropy: Optimal Noise Allocation for Diffusion Training in the Convex Regime
Luca Ambrogioni, Giulio Franzese, Alberto Foresti +7
How should a diffusion model decide which noise levels to train on, and how much? Despite the importance of this choice, current noise schedules are based largely on heuristics or…
Towards Diverse and Comprehensive Benchmarks for Mutual Information Estimation
Alberto Foresti, Ivan Butakov, Alexander Tolmachev +3
Mutual information (MI) estimation is a central problem in machine learning and statistics; however, existing benchmarks typically evaluate estimators on simplified, low-dimensiona…
Improved Sampling Schedules for Discrete Diffusion Models
Alberto Foresti, Mustapha Bounoua, Giulio Franzese +2
Discrete diffusion models have emerged as a powerful paradigm for generative modeling on sequence data; however, the information-theoretic principles governing their reverse proces…
INFO-SEDD: Continuous Time Markov Chains as Scalable Information Metrics Estimators
Alberto Foresti, Giulio Franzese, Pietro Michiardi
Information-theoretic quantities play a crucial role in understanding non-linear relationships between random variables and are widely used across scientific disciplines. However,…
Latent Abstractions in Generative Diffusion Models
Giulio Franzese, Mattia Martini, Giulio Corallo +2
In this work we study how diffusion-based generative models produce high-dimensional data, such as an image, by implicitly relying on a manifestation of a low-dimensional set of la…
MINDE: Mutual Information Neural Diffusion Estimation
Giulio Franzese, Mustapha Bounoua, Pietro Michiardi
In this work we present a new method for the estimation of Mutual Information (MI) between random variables. Our approach is based on an original interpretation of the Girsanov the…
Multi-modal Latent Diffusion
Mustapha Bounoua, Giulio Franzese, Pietro Michiardi
Multi-modal data-sets are ubiquitous in modern applications, and multi-modal Variational Autoencoders are a popular family of models that aim to learn a joint representation of the…
One-Line-of-Code Data Mollification Improves Optimization of Likelihood-based Generative Models
Ba-Hien Tran, Giulio Franzese, Pietro Michiardi +1
Generative Models (GMs) have attracted considerable attention due to their tremendous success in various domains, such as computer vision where they are capable to generate impress…
Continuous-Time Functional Diffusion Processes
Giulio Franzese, Giulio Corallo, Simone Rossi +3
We introduce Functional Diffusion Processes (FDPs), which generalize score-based diffusion models to infinite-dimensional function spaces. FDPs require a new mathematical framework…
How Much is Enough? A Study on Diffusion Times in Score-based Generative Models
Giulio Franzese, Simone Rossi, Lixuan Yang +4
Score-based diffusion models are a class of generative models whose dynamics is described by stochastic differential equations that map noise into data. While recent works have sta…
Revisiting the Effects of Stochasticity for Hamiltonian Samplers
Giulio Franzese, Dimitrios Milios, Maurizio Filippone +1
We revisit the theoretical properties of Hamiltonian stochastic differential equations (SDES) for Bayesian posterior sampling, and we study the two types of errors that arise from…
Isotropic SGD: a Practical Approach to Bayesian Posterior Sampling
Giulio Franzese, Rosa Candela, Dimitrios Milios +2
In this work we define a unified mathematical framework to deepen our understanding of the role of stochastic gradient (SG) noise on the behavior of Markov chain Monte Carlo sampli…
Sparsification as a Remedy for Staleness in Distributed Asynchronous SGD
Rosa Candela, Giulio Franzese, Maurizio Filippone +1
Large scale machine learning is increasingly relying on distributed optimization, whereby several machines contribute to the training process of a statistical model. In this work w…
Deep Information Networks
Giulio Franzese, Monica Visintin
We describe a novel classifier with a tree structure, designed using information theory concepts. This Information Network is made of information nodes, that compress the input dat…