no code implementations • 8 Apr 2024 • Seyedehdelaram Esfahani, Giovanni De Toni, Bruno Lepri, Andrea Passerini, Katya Tentori, Massimo Zancanaro
Our results suggest that users may recognize that the guided interaction paradigm improves efficiency.
no code implementations • 4 Apr 2024 • Marco Bronzini, Carlo Nicolini, Bruno Lepri, Jacopo Staiano, Andrea Passerini
We propose an end-to-end framework that jointly decodes the factual knowledge embedded in the latent space of LLMs from a vector space to a set of ground predicates and represents its evolution across the layers using a temporal knowledge graph.
1 code implementation • 29 Mar 2024 • Burcu Sayin, Pasquale Minervini, Jacopo Staiano, Andrea Passerini
We explore the potential of Large Language Models (LLMs) to assist and potentially correct physicians in medical decision-making tasks.
no code implementations • 25 Mar 2024 • Debodeep Banerjee, Stefano Teso, Burcu Sayin, Andrea Passerini
As a remedy, we introduce learning to guide (LTG), an alternative framework in which - rather than taking control from the human expert - the machine provides guidance useful for decision making, and the human is entirely responsible for coming up with a decision.
1 code implementation • 19 Feb 2024 • Emanuele Marconato, Samuele Bortolotti, Emile van Krieken, Antonio Vergari, Andrea Passerini, Stefano Teso
Neuro-Symbolic (NeSy) predictors that conform to symbolic knowledge - encoding, e. g., safety constraints - can be affected by Reasoning Shortcuts (RSs): They learn concepts consistent with the symbolic knowledge by exploiting unintended semantics.
no code implementations • 7 Feb 2024 • Paolo Morettin, Andrea Passerini, Roberto Sebastiani
The probabilistic formal verification (PFV) of AI systems is in its infancy.
1 code implementation • 9 Oct 2023 • Marco Bronzini, Carlo Nicolini, Bruno Lepri, Andrea Passerini, Jacopo Staiano
This poses a challenge in efficiently gathering and aligning the data into a unified framework to derive insights related to Corporate Social Responsibility (CSR).
1 code implementation • 29 Sep 2023 • Francesco Ferrini, Antonio Longa, Andrea Passerini, Manfred Jaeger
Existing multi-relational graph neural networks use one of two strategies for identifying informative relations: either they reduce this problem to low-level weight learning, or they rely on handcrafted chains of relational dependencies, called meta-paths.
no code implementations • 14 Sep 2023 • Emanuele Marconato, Andrea Passerini, Stefano Teso
This allows us to derive a principled notion of alignment between the machine representation and the vocabulary of concepts understood by the human.
no code implementations • 11 Aug 2023 • Debodeep Banerjee, Stefano Teso, Andrea Passerini
In learning to defer, a predictor identifies risky decisions and defers them to a human expert.
1 code implementation • NeurIPS 2023 • Emanuele Marconato, Stefano Teso, Antonio Vergari, Andrea Passerini
Neuro-Symbolic (NeSy) predictive models hold the promise of improved compliance with given constraints, systematic generalization, and interpretability, as they allow to infer labels that are consistent with some prior knowledge by reasoning over high-level concepts extracted from sub-symbolic inputs.
no code implementations • 9 May 2023 • Luca Erculiani, Andrea Bontempelli, Andrea Passerini, Fausto Giunchiglia
We achieve this goal by implementing an algorithm which, for any object, recursively recognizes its visual genus and its visual differentia.
1 code implementation • 31 Mar 2023 • Samy Badreddine, Gianluca Apriceno, Andrea Passerini, Luciano Serafini
In this paper, we introduce Interval Real Logic (IRL), a two-sorted logic that interprets knowledge such as sequential properties (traces) and event properties using sequences of real-featured data.
no code implementations • 22 Mar 2023 • Emanuele Marconato, Stefano Teso, Andrea Passerini
This setup offers clear advantages in terms of consistency to symbolic prior knowledge, and is often believed to provide interpretability benefits in that - by virtue of complying with the knowledge - the learned concepts can be better understood by human stakeholders.
no code implementations • 13 Feb 2023 • Giuseppe Spallitta, Gabriele Masina, Paolo Morettin, Andrea Passerini, Roberto Sebastiani
The development of efficient exact and approximate algorithms for probabilistic inference is a long-standing goal of artificial intelligence research.
1 code implementation • 2 Feb 2023 • Emanuele Marconato, Gianpaolo Bontempo, Elisa Ficarra, Simone Calderara, Andrea Passerini, Stefano Teso
We introduce Neuro-Symbolic Continual Learning, where a model has to solve a sequence of neuro-symbolic tasks, that is, it has to map sub-symbolic inputs to high-level concepts and compute predictions by reasoning consistently with prior knowledge.
no code implementations • 2 Feb 2023 • Antonio Longa, Veronica Lachi, Gabriele Santin, Monica Bianchini, Bruno Lepri, Pietro Lio, Franco Scarselli, Andrea Passerini
Graph Neural Networks (GNNs) have become the leading paradigm for learning on (static) graph-structured data.
2 code implementations • 27 Oct 2022 • Antonio Longa, Steve Azzolin, Gabriele Santin, Giulia Cencetti, Pietro Liò, Bruno Lepri, Andrea Passerini
Following a fast initial breakthrough in graph based learning, Graph Neural Networks (GNNs) have reached a widespread application in many science and engineering fields, prompting the need for methods to understand their decision process.
1 code implementation • 13 Oct 2022 • Steve Azzolin, Antonio Longa, Pietro Barbiero, Pietro Liò, Andrea Passerini
While instance-level explanation of GNN is a well-studied problem with plenty of approaches being developed, providing a global explanation for the behaviour of a GNN is much less explored, despite its potential in interpretability and debugging.
no code implementations • 30 Sep 2022 • Burcu Sayin, Fabio Casati, Andrea Passerini, Jie Yang, Xinyue Chen
In this paper, we argue that the way we have been training and evaluating ML models has largely forgotten the fact that they are applied in an organization or societal context as they provide value to people.
1 code implementation • 28 Jun 2022 • Giuseppe Spallitta, Gabriele Masina, Paolo Morettin, Andrea Passerini, Roberto Sebastiani
Weighted Model Integration (WMI) is a popular formalism aimed at unifying approaches for probabilistic inference in hybrid domains, involving logical and algebraic constraints.
1 code implementation • 31 May 2022 • Emanuele Marconato, Andrea Passerini, Stefano Teso
There is growing interest in concept-based models (CBMs) that combine high-performance and interpretability by acquiring and reasoning with a vocabulary of high-level concepts.
1 code implementation • 31 May 2022 • Andrea Bontempelli, Stefano Teso, Katya Tentori, Fausto Giunchiglia, Andrea Passerini
We propose ProtoPDebug, an effective concept-level debugger for ProtoPNets in which a human supervisor, guided by the model's explanations, supplies feedback in the form of what part-prototypes must be forgotten or kept, and the model is fine-tuned to align with this supervision.
1 code implementation • 27 May 2022 • Giovanni De Toni, Paolo Viappiani, Stefano Teso, Bruno Lepri, Andrea Passerini
It is paramount that the sequence of actions does not require too much effort for users to implement.
no code implementations • 20 May 2022 • Stefano Teso, Laurens Bliek, Andrea Borghesi, Michele Lombardi, Neil Yorke-Smith, Tias Guns, Andrea Passerini
The challenge is to learn them from available data, while taking into account a set of hard constraints that a solution must satisfy, and that solving the optimisation problem (esp.
no code implementations • 10 May 2022 • Andrea Bontempelli, Marcelo Rodas Britez, Xiaoyue Li, Haonan Zhao, Luca Erculiani, Stefano Teso, Andrea Passerini, Fausto Giunchiglia
We focus on the development of AIs which live in lifelong symbiosis with a human.
1 code implementation • 18 Jan 2022 • Giovanni De Toni, Bruno Lepri, Andrea Passerini
Being able to provide counterfactual interventions - sequences of actions we would have had to take for a desirable outcome to happen - is essential to explain how to change an unfavourable decision by a black-box machine learning model (e. g., being denied a loan request).
no code implementations • 11 Nov 2021 • Burcu Sayin, Jie Yang, Andrea Passerini, Fabio Casati
We motivate why the science of learning to reject model predictions is central to ML, and why human computation has a lead role in this effort.
no code implementations • 23 Sep 2021 • Andrea Bontempelli, Fausto Giunchiglia, Andrea Passerini, Stefano Teso
In this paper, we tackle interactive debugging of "gray-box" concept-based models (CBMs).
no code implementations • NeurIPS Workshop AIPLANS 2021 • Giovanni De Toni, Luca Erculiani, Andrea Passerini
We showcase the potential of our approach by learning the Quicksort algorithm, showing how the ability to deal with arguments is crucial for learning and generalization.
1 code implementation • NeurIPS 2021 • Stefano Teso, Andrea Bontempelli, Fausto Giunchiglia, Andrea Passerini
We tackle sequential learning under label noise in applications where a human supervisor can be queried to relabel suspicious examples.
no code implementations • 26 Apr 2021 • Fausto Giunchiglia, Luca Erculiani, Andrea Passerini
In this paper we provide a theory and an algorithm for how to build substance concepts which are in a one-to-one correspondence with classifications concepts, thus paving the way to the seamless integration between natural language descriptions and visual perception.
no code implementations • 31 Mar 2021 • Paolo Dragone, Stefano Teso, Andrea Passerini
We propose Nester, a method for injecting neural networks into constrained structured predictors.
1 code implementation • 27 Mar 2021 • Andrea Bontempelli, Fausto Giunchiglia, Andrea Passerini, Stefano Teso
Motivated by this, we introduce TRCKD, a novel approach that combines automated drift detection and adaptation with an interactive stage in which the user is asked to disambiguate between different kinds of KD.
1 code implementation • 28 Jan 2021 • Alessia Bertugli, Stefano Vincenzi, Simone Calderara, Andrea Passerini
Future deep learning systems call for techniques that can deal with the evolving nature of temporal data and scarcity of annotations when new problems occur.
1 code implementation • 15 Dec 2020 • Giovanni Pellegrini, Alessandro Tibo, Paolo Frasconi, Andrea Passerini, Manfred Jaeger
Learning on sets is increasingly gaining attention in the machine learning community, due to its widespread applicability.
1 code implementation • 2 Nov 2020 • Andrea Bontempelli, Stefano Teso, Fausto Giunchiglia, Andrea Passerini
The ability to learn from human supervision is fundamental for personal assistants and other interactive applications of AI.
1 code implementation • 17 Sep 2020 • Alessia Bertugli, Stefano Vincenzi, Simone Calderara, Andrea Passerini
In real-world applications, data do not reflect the ones commonly used for neural networks training, since they are usually few, unlabeled and can be available as a stream.
1 code implementation • NeurIPS 2020 • Luca Di Liello, Pierfrancesco Ardino, Jacopo Gobbi, Paolo Morettin, Stefano Teso, Andrea Passerini
Generative Adversarial Networks (GANs) struggle to generate structured objects like molecules and game maps.
no code implementations • 26 Jun 2020 • Stefano Mandelli, Elenia Manzan, Aniello Mennella, Francesco Cavaliere, Daniele Viganò, Cristian Franceschet, Paolo de Bernardis, Marco Bersanelli, Maria Gabriella Castellano, Alessandro Coppolecchia, Angelo Cruciani, Massimo Gervasi, Luca Lamagna, Andrea Limonta, Silvia Masi, Alessandro Paiella, Andrea Passerini, Giorgio Pettinari, Francesco Piacentini, Elisabetta Tommasi, Angela Volpe, Mario Zannoni
We present the design, manufacturing, and testing of a 37-element array of corrugated feedhorns for Cosmic Microwave Background (CMB) measurements between $140$ and $170$ GHz.
Instrumentation and Methods for Astrophysics
1 code implementation • 6 Dec 2019 • Luca Erculiani, Fausto Giunchiglia, Andrea Passerini
We present a framework capable of tackilng the problem of continual object recognition in a setting which resembles that under whichhumans see and learn.
1 code implementation • 22 Nov 2017 • Paolo Dragone, Stefano Teso, Mohit Kumar, Andrea Passerini
We propose a decomposition technique for large preference-based decision problems relying exclusively on inference and feedback over partial configurations.
1 code implementation • 21 Nov 2017 • Paolo Dragone, Stefano Teso, Andrea Passerini
The preferences are typically learned by querying the user for choice feedback over pairs or sets of objects.
no code implementations • 6 Dec 2016 • Stefano Teso, Paolo Dragone, Andrea Passerini
When faced with complex choices, users refine their own preference criteria as they explore the catalogue of options.
no code implementations • 17 Jun 2016 • Seyed Mostafa Kia, Andrea Passerini
Despite extensive studies of this type, at present, there is no formal definition for interpretability of brain decoding models.
1 code implementation • 20 Apr 2016 • Stefano Teso, Andrea Passerini, Paolo Viappiani
In this paper we propose an approach to preference elicitation that is suitable to large configuration spaces beyond the reach of existing state-of-the-art approaches.
no code implementations • 18 Aug 2015 • Paolo Campigotto, Roberto Battiti, Andrea Passerini
CLEO iteratively alternates a preference elicitation step, where pairs of candidate solutions are selected based on the current utility model, and a refinement step where the utility is refined by incorporating the feedback received.
1 code implementation • 7 May 2014 • Stefano Teso, Roberto Sebastiani, Andrea Passerini
The main idea is to leverage a state-of-the-art generalized Satisfiability Modulo Theory solver for implementing the inference and separation oracles of Structured Output SVMs.
no code implementations • 18 Feb 2014 • Stefano Teso, Roberto Sebastiani, Andrea Passerini
Generally speaking, the goal of constructive learning could be seen as, given an example set of structured objects, to generate novel objects with similar properties.
no code implementations • NeurIPS 2008 • Paolo Frasconi, Andrea Passerini
Metal binding is important for the structural and functional characterization of proteins.