Italy

In recent years, there have been many attempts to combine XAI with the field of symbolic AI in order to generate explanations for neural networks that are more interpretable and better align with human reasoning, with one prominent candidate for this synergy being the sub-field of computational argumentation. One method is to represent neural networks with quantitative bipolar argumentation frameworks (QBAFs) equipped with a particular semantics. The resulting QBAF can then be viewed as an explanation for the associated neural network. In this paper, we explore a novel LRP-based semantics under a new QBAF variant, namely neural QBAFs (nQBAFs). Since an nQBAF of a neural network is typically large, the nQBAF must be simplified before being used as an explanation. Our empirical evaluation indicates that the manner of this simplification is all important for the quality of the resulting explanation.

AIxIA 2021

LRP-Based Argumentative Explanations for Neural Networks

workshop paper

[![](https://assets.underline.io/uploads/markdown_image/1/image/9a88ae95fbde06e80f4bcf6d4f7d133a.png)](https://certificateaixia.paperform.co/)
The certificate will be provided by December 17. 
In case you need it urgently please contact: francesca.gasparini@unimib.it. 

 

The 20th International Conference of the Italian Association for Artificial Intelligence (**AIxIA 2021**) is organized by AIxIA (Associazione Italiana per l’Intelligenza Artificiale), which is a non-profit scientific society founded in 1988 and devoted to the promotion of Artificial Intelligence. The society aims to increase the public awareness of AI, encourage the teaching of it and promote research in the field.

 
 



**General Chair**

Stefania Bandini, University of Milano-Bicocca, Italyand RCAST - The University of Tokyo, Japan


**Program Chairs**

Viviana Mascardi, University of Genova, Italy; Matteo Palmonari, University of Milano-Bicocca, Italy; Giuseppe Vizzari, University of Milano-Bicocca, Italy

**Organizing Committee**

Francesca Gasparini, University of Milano-Bicocca, Italy (**Chair**); Stefano Ferilli, University of Bari, Italy; Marta Giltri, University of Milano-Bicocca, Italy; Daniela Briola, University of Insubria, Italy


**Publicity and Communication Committee**

Elisabetta Fersini, University of Milano-Bicocca, Italy (**Chair**); Angelo Ferrando, University of Genova, Italy; Francesca Alessandra Lisi, University of Bari, Italy; Giovanni Stilo, University of L'Aquila, Italy; Marco Viviani, University of Milano-Bicocca, Italy


**Workshop Chairs**

Gabriella Cortellessa, ISTC-CNR, Rome, Italy; Silvia Rossi, University of Naples Federico II, Italy; Fabio Stella, University of Milano-Bicocca, Italy


**PC Members**

Fulvio Mastrogiovanni, University of Genova, Italy; Roberto Basili, University of Rome Tor Vergata, Italy; Nicola Gatti, Polytechnic University of Milan, Italy; Stefano Bistarelli, University of Perugia, Italy; Piero Poccianti, Consorzio Operativo Gruppo MPS, Italy; Francesco Scarcello, University of Calabria, Italy; Flavio S. Correa Da Silva, University of São Paulo, Brazil; Elena Cabrio, Université Côte d'Azur, CNRS, Inria, I3S, France; Antonio Lieto, University of Turin, Italy; Diego Calvanese, Free University of Bozen-Bolzano, Italy; Alessio Micheli, University of Pisa, Italy; Francesca A. Lisi University of Bari Aldo Moro, Italy; Mauro Vallati, University of Huddersfield, United Kingdom; Bernardo Magnini, Fondazione Bruno Kessler (FBK), Italy; Luigia C. Aiello, Sapienza University of Rome, Italy; Rafael H. Bordini, Pontifical Catholic University of Rio Grande do Sul, Brazil; Chiara Ghidini, Fondazione Bruno Kessler (FBK), Italy; Federico Bergenti, University of Parma, Italy; Francesco Ricca, University of Calabria, Italy; Andrea Burattin, Technical University of Denmark, Denmark; Angelo Oddi, ISTC-CNR, Italian National Research Council, Italy; Farshad Badie, Aalborg University, Denmark; Emanuel Sallinger, University of Oxford, United Kingdom; Marin Lujak, University Rey Juan Carlos, Spain; Giancarlo Fortino, University of Calabria, Italy; Salvatore Gaglio, University of Palermo, Italy; Johannes K. Fichte, TU Dresden, Germany; Matteo Baldoni, University of Turin, Italy; Fabrizio Riguzzi, Univeristy of Ferrara, Italy; Ivana Dusparic, Trinity College Dublin, Ireland; Serena Villata, CNRS - Laboratoire d'Informatique, Signaux et Systèmes de Sophia-Antipolis, France; Paolo Rosso, Technical University of Valencia, Spain; Ana L. C. Bazzan, Federal University of Rio Grande do Sul, Brazil; Gianluigi Greco, University of Calabria, Italy; Stefano Cagnoni, University of Parma, Italy; Gian Luca Pozzato, University of Turin, Italy; Salvatore Ruggieri, University of Turin, Italy; Claudio Di Ciccio, Sapienza University of Rome, Italy; Eloisa Vargiu, Cetaqua, Water Technology Center, Spain; Marco Maratea, University of Genova, Italy; Andrea Roli, University of Bologna, Italy; Viviana Patti, University of Turin, Italy; Stefania Costantini, University of L'Aquila, Italy; Antonius Weinzierl, Vienna University of Technology, Austria; Silvia Rossi, University of Naples Federico II, Italy; Giovanni Semeraro, University of Bari, Italy; Luciano Serafini, Fondazione Bruno Kessler (FBK), Italy; Laura Giordano, University of Eastern Piedmont, Italy; Ernesto Jimenez-Ruiz, City, University of London, United Kingdom; Marco Gavanelli, University of Ferrara, Italy; Federico Chesani, University of Bologna, Italy

Please register with AIxIA to be able to enter the conference on Underline.io. If you have already registered, please login to Underline.io to be able to see full conference content. In case you do not have an Underline account, please create one. Thank you

Please login to your Underline account

The 20th International Conference of the Italian Association for Artificial Intelligence (AIxIA 2021) is organized by AIxIA (Associazione Italiana per l’Intelligenza Artificiale), which is a non-profit scientific society founded in 1988 and devoted to the promotion of Artificial Intelligence. The society aims to increase the public awareness of AI, encourage the teaching of it and promote research in the field.

In this paper, we present BERT-POS, a simple method for encoding syntax into BERT embeddings without retraining or finetuning data, based on Part-Of-Speech (POS). Although fine-tuning is the most popular method to apply BERT models on domain datasets, it remains expensive in terms of training time, computing resources, training data selection and retraining frequency. Our alternative works at the preprocessing level and relies on POS tagging sentences. It gives interesting results for words similarity regarding out-of-vocabulary both in terms of domain-specific words and misspellings. More specifically, the experiments were done on French language, but we believe that they would be similar on others.

Easy-to-use combination of POS and BERT model for domain-specific and misspelled terms

Artificial neural networks for source code processing are often able to develop rich representations for different program properties. In this software demo, we exploit the knowledge learned by a neural network, i.e. the activation of a chosen neuron, for guiding the evolution of programs that satisfy given requirements.

Integrating Grammatical Evolution with Neural Fitness Functions

In this paper I will tackle the debated question of whether Deep Neural Networks “understand” Natural Language when they process it either for classification or for generation tasks. I will start by quoting work currently carried out in the field of explainable AI then I will look into the internal procedures adopted by current DNNs to cope with the enormous and untreatable dimension of vocabulary size due to the presence of rare words and the use of subwords or n-gram character sequences. To explain the inability of DNNs to reuse previously discovered knowledge I will deal with the issue of generalization and the lack of systematicity in learning models. I will also propose reasons to motivate the difficulty of generalizing with morphology-rich and weakly configurational languages like Italian. In particular vocabulary size and rare words need a different approach from the one used with English texts. Eventually I will propose some alternative ways to deal with the problems discussed in the paper

What's Wrong with Deep Learning for Meaning Understanding

Shapley values has established itself as one of the most appropriate and theoretically sound frameworks for explaining predictions from complex machine learning models. The popularity of Shapley values in the explanation setting is probably due to Shapley values’ unique theoretical properties. The main drawback with Shapley values, however, is that the computational complexity grows exponentially in the number of input features, making it unfeasible in many real world situations where there could be hundreds or thousands of features. Furthermore, with many (dependent) features, presenting/visualizing and interpreting the computed Shapley values also become challenging. The present paper introduces and showcases a method that we call groupShapley. The idea of the method is to group features and then compute and present Shapley values for these groups instead of for all individual features. Reducing hundreds or thousands of features to half a dozen or so feature groups makes precise computations practically feasible, and the presentation and knowledge extraction greatly simplified. We give practical advice for using the approach and illustrate its usability in three different real world examples. The examples vary in both data type (regular tabular data and time series), feature dimension (medium to high), and application (insurance, genetics, and banking).

Efficient and simple prediction explanations with groupShapley: A practical perspective

Disagreement in annotation, traditionally treated mostly as noise, is now more and more often considered as a source of valuable information instead. We investigate a particular form of disagreement, occurring when the focus of an annotated dataset is a subjective and controversial phenomenon, therefore inducing a certain degree of polarization among the annotators’ judgments. We argue that the polarization is indicative of the conflicting perspectives held by different annotator groups, and propose a quantitative method to model this phenomenon. Moreover, we introduce a method to automatically identify shared perspectives stemming from a common background. We test our method on several corpora in English and Italian, manually annotated according to their hate speech content, validating prior knowledge about the groups of annotators, when available, and discovering characteristic traits among annotators with unknown background. We found several precisely defined perspectives, described in terms of increased sensitivity towards textual content expressing attitudes such as xenophobia, islamophobia, and homophobia.

Mining Annotator Perspectives from Hate Speech Corpora

We consider the optimization problem of constructing a binary orthogonal array (OA) starting from a bigger one, by removing a specified amount of lines. In particular, we develop a genetic algorithm (GA) where the underlying chromosomes are constant-weight binary strings that specify the lines to be cancelled from the starting OA. Such chromosomes are then evolved through balanced crossover and mutation operators to preserve the number of ones in them. The fitness function evaluates the matrices obtained from these chromosomes by measuring their distance from satisfying the constraints of an OA smaller than the starting one. We perform a preliminary experimental validation of the proposed genetic algorithm by crafting the initial OA as a random permutation of several blocks of the basic parity-check array, thereby guaranteeing the existence of an optimal solution.

Deriving Smaller Orthogonal Arrays from Bigger Ones with Genetic Algorithms

On the Applications of Evolutionary Algorithms to the Cybersecurity Domain

Digital images made a big difference in everyday life and science. They became crucial part in various areas such as medicine, astronomy, agriculture, etc. A common problem in applications with digital images is classification. Since recently, convolutional neural network has been widely used for image classification and classification accuracy is dramatically improved. To have the best possible classification accuracy of a CNN for a specific application, the necessary step is tuning various hyperparameters such as the number of different layers, number of neurons in each layer, optimization algorithm, activation functions, kernel size, optimization algorithm, etc. Recent studies show that swarm intelligence algorithms used for CNN’s hyperparameters tuning give promising results and further research is needed. In this paper, a short analysis of swarm intelligence algorithms and their application to tuning CNN hyperparameters is presented.

LRP-Based Argumentative Explanations for Neural Networks

Downloads

Next from AIxIA 2021

Easy-to-use combination of POS and BERT model for domain-specific and misspelled terms

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

LRP-Based Argumentative Explanations for Neural Networks

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from AIxIA 2021

Easy-to-use combination of POS and BERT model for domain-specific and misspelled terms

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Downloads