Giorgio Giannone

I am a PhD student at the Section for Cognitive Systems at the Technical University of Denmark (DTU), supervised by Ole Winther and Søren Hauberg.

CV  /  Scholar  /  Github  /  LinkedIn

I am broadly interested in Probabilistic Machine Learning, Perception and Geometry,
with a focus on Conditional Generative Models, Hierarchical Variational Inference,
Few-Shot Generation, Transfer Learning, and Diffusion Models.

Research

I am interested in the generalization and adaptation capacities of hierarchical generative models.

Large generative models trained on millions of data points exhibit emergent adaptation properties like in-context learning, being able to solve novel tasks given a handful of samples. However, such adaptation properties emerge only with scale, being absent for middle-size models trained on small datasets, a typical scenario in engineering design and scientific discovery, where data collection is expensive and computational constraints are present.

My research goal is to bridge the gap in adaptation capabilities between large and middle-size generative models. Leveraging the tools of hierarchical inference and using the training dataset in such a way as to encourage adaptation at inference time, we can learn expressive families of conditional generative processes and perform few-shot generation and transfer learning in the small data regime.

Papers
blind-date

Few-Shot Diffusion Models
Giorgio Giannone, Didrik Nielsen, Ole Winther
under-review, 2022

Denoising diffusion probabilistic models (DDPM) are powerful hierarchical latent variable models with remarkable sample generation quality and training stability. These properties can be attributed to parameter sharing in the generative hierarchy, as well as a parameter-free diffusion-based inference procedure. In this paper, we present Few-Shot Diffusion Models (FSDM), a framework for few-shot generation leveraging conditional DDPMs. FSDMs are trained to adapt the generative process conditioned on a small set of images from a given class by aggregating image patch information using a set-based Vision Transformer (ViT). At test time, the model is able to generate samples from previously unseen classes conditioned on as few as 5 samples from that class. We empirically show that FSDM can perform few-shot generation and transfer to new datasets taking full advantage of the conditional DDPM.

blind-date

SCHA-VAE: Hierarchical Context Aggregation for Few-Shot Generation
Giorgio Giannone, Ole Winther
International Conference on Machine Learning, ICML, 2022

A few-shot generative model should be able to generate data from a novel distribution by only observing a limited set of examples. In few-shot learning the model is trained on data from many sets from distributions sharing some underlying properties such as sets of characters from different alphabets or objects from different categories. We extend current latent variable models for sets to a fully hierarchical approach with an attention-based point to set-level aggregation and call our method SCHA-VAE for Set-Context-Hierarchical-Aggregation Variational Autoencoder. We explore likelihood-based model comparison, iterative data sampling, and adaptation-free out-of-distribution generalization. Our results show that the hierarchical formulation better captures the intrinsic variability within the sets in the small data regime. This work generalizes deep latent variable approaches to few-shot learning, taking a step toward large-scale few-shot generation with a formulation that readily works with current state-of-the-art deep generative models.

blind-date

Just Mix Once: Worst-group Generalization by Group Interpolation
Giorgio Giannone, Serhii Havrylov, Jordan Massiah, Emine Yilmaz, Yunlong Jiao
Under-review, 2022 - also Distribution Shifts Workshop, NeurIPS, 2021

Advances in deep learning theory have revealed how average generalization relies on superficial patterns in data. The consequences are brittle models with poor performance with shift in group distribution at test time. When group annotation is available, we can use robust optimization tools to tackle the problem. However, identification and annotation are time-consuming, especially on large datasets. A recent line of work leverages self-supervision and oversampling to improve generalization on minority groups without group annotation. We propose to unify and generalize these approaches using a class-conditional variant of mixup tailored for worst-group generalization. Our approach, Just Mix Once (JM1), interpolates samples during learning, augmenting the training distribution with a continuous mixture of groups. JM1 is domain agnostic and computationally efficient, can be used with any level of group annotation, and performs on par or better than the state-of-the-art on worst-group generalization. Additionally, we provide a simple explanation of why JM1 works.

blind-date

Hierarchical Few-Shot Generative Models
Giorgio Giannone, Ole Winther
Meta-Learning Workshop, NeurIPS, 2021

A few-shot generative model should be able to generate data from a distribution by only observing a limited set of examples. In few-shot learning the model is trained on data from many sets from different distributions sharing some underlying properties such as sets of characters from different alphabets or sets of images of different type objects. We study a latent variables approach that extends the Neural Statistician to a fully hierarchical approach with an attention-based point to set-level aggregation. We extend the previous work to iterative data sampling, likelihood-based model comparison, and adaptation-free out of distribution generalization. Our results show that the hierarchical formulation better captures the intrinsic variability within the sets in the small data regime. With this work we generalize deep latent variable approaches to few-shot learning, taking a step towards large-scale few-shot generation with a formulation that readily can work with current state-of-the-art deep generative models.

blind-date

Transformation-aware Variational Autoencoder
Giorgio Giannone, Saeed Saremi, Jonathan Masci, Christian Osendorfer
Technical Report, 2020

We extend the framework of variational autoencoders to represent transformations explicitly in the latent space. This is achieved in the form of a generative model structured such that the group of transformations that act in the input space is instead represented by latent variables which are linear operators that only act in the latent space. In the family of hierarchical graphical models that emerges, the latent space is populated by higher order objects which are inferred jointly with the latent representations they act on.

blind-date

Real-time Classification from Short Event-Camera Streams using Input-filtering Neural ODEs
Giorgio Giannone, Asha Anoosheh, Alessio Quaglino, Pierluca D'Oro, Marco Gallieri, Jonathan Masci
Interpretable Inductive Biases and Physically Structured Learning Workshop, NeurIPS, 2020

Event-based cameras are novel, efficient sensors inspired by the human vision system, generating an asynchronous, pixel-wise stream of data. Learning from such data is generally performed through heavy preprocessing and event integration into images. This requires buffering of possibly long sequences and can limit the response time of the inference system. In this work, we instead propose to directly use events from a DVS camera, a stream of intensity changes and their spatial coordinates. This sequence is used as the input for a novel asynchronous RNN-like architecture, the Input-filtering Neural ODEs.

blind-date

No Representation without Transformation
Giorgio Giannone, Jonathan Masci, Christian Osendorfer
Bayesian Deep Learning and Perception as Generative Reasoning Workshops, NeurIPS , 2019

We propose to extend Latent Variable Models with a simple idea: learn to encode not only samples but also transformations of such samples. This means that the latent space is not only populated by embeddings but also by higher order objects that map between these embeddings. We show how a hierarchical graphical model can be utilized to enforce desirable algebraic properties of such latent mappings.

blind-date

Learning Common Representation from RGB and Depth Images
Giorgio Giannone, Boris Chidlovskii
Multimodal Learning and Applications Workshop, CVPR, 2019

We propose a new deep learning architecture for the tasks of semantic segmentation and depth prediction from RGB-D images. We revise the state of art based on the RGB and depth feature fusion, where both modalities are assumed to be available at train and test time. We propose a new architecture where the feature fusion is replaced with a common deep representation. Combined with an encoder-decoder type of the network, the architecture can jointly learn models for semantic segmentation and depth estimation based on their common representation.

Open-source
blind-date

GT4SD: Generative Toolkit for Scientific Discovery
GT4SD Team, 2022

The GT4SD (Generative Toolkit for Scientific Discovery) is an open-source platform to accelerate hypothesis generation in the scientific discovery process. It provides a library for making state-of-the-art generative AI models easier to use.

Patents
blind-date

Method and apparatus for semantic segmentation and depth completion using a convolutional neural network
Boris Chidlovskii, Giorgio Giannone
US Patent US11263756B2, 2022

Theses
blind-date

Learning Common Representation for Scene Understanding
Giorgio Giannone
Master's Thesis, Data Science, Sapienza University of Rome, 2018

blind-date

Bubble Dynamics in Turbulent Shear Flows
Giorgio Giannone
Master's Thesis, Mechanical Engineering, Sapienza University of Rome, 2016


the original