Italy

The recent success of deep learning models in solving complex problems and in different domains has increased interest in understanding what they learn. Therefore, different approaches have been employed to explain these models, one of which uses human-understandable concepts as explanations. Two examples of methods that use this approach are Network Dissection and Compositional explanations. The former explains units using atomic concepts, while the latter makes explanations more expressive, replacing atomic concepts with logical forms. While intuitively, logical forms are more informative than atomic concepts, it is not clear how to quantify this improvement, and their evaluation is often based on the same metric that is optimized during the search-process and on the usage of hyper-parameters to be tuned. In this paper, we propose to use as evaluation metric the Detection Accuracy, which measures units&#39; consistency of detection of their assigned explanations. We show that this metric (1) evaluates explanations of different lengths effectively, (2) can be used as a stopping criterion for the compositional explanation search, eliminating the explanation length hyper-parameter, and (3) exposes new specialized units whose length 1 explanations are the perceptual abstractions of their longer explanations.

AIxIA 2021

Detection Accuracy for Evaluating Compositional Explanations of Units

[![](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.

The recent success of deep learning models in solving complex problems and in different domains has increased interest in understanding what they learn. Therefore, different approaches have been employed to explain these models, one of which uses human-understandable concepts as explanations. Two examples of methods that use this approach are Network Dissection and Compositional explanations. The former explains units using atomic concepts, while the latter makes explanations more expressive, replacing atomic concepts with logical forms. While intuitively, logical forms are more informative than atomic concepts, it is not clear how to quantify this improvement, and their evaluation is often based on the same metric that is optimized during the search-process and on the usage of hyper-parameters to be tuned. In this paper, we propose to use as evaluation metric the Detection Accuracy, which measures units' consistency of detection of their assigned explanations. We show that this metric (1) evaluates explanations of different lengths effectively, (2) can be used as a stopping criterion for the compositional explanation search, eliminating the explanation length hyper-parameter, and (3) exposes new specialized units whose length 1 explanations are the perceptual abstractions of their longer explanations.

technical paper

Explainable Artificial Intelligence for Technology Policy Making Using Attribution Networks

Neural Machine Translation (NMT) systems have been shown to perform impressively on many language pairs compared to Statistical Machine Translation (SMT). However, these systems are data-intensive, which is problematic for the majority of language pairs, and especially for low-resource languages. In this work, we address this issue in the case of certain Arabic dialects, those variants of Modern Standard Arabic (MSA) that are spelling non-standard, morphologically rich, and yet resource-poor variants. Here, we have experimented with several multitasking learning strategies to take advantage of the relationships between these dialects. Despite the simplicity of this idea, empirical results show that several multitasking learning strategies are capable of achieving remarkable performance compared to statistical machine translation. For instance, we obtained the BLUE scores for the Algerian → Modern-Standard-Arabic and the Moroccan → Palestinian of 35.06 and 27.55, respectively, while the scores obtained with a statistical method are 15.1 and 18.91 respectively. We show that on 42 machine translation experiments, and despite the use of a small corpus, multitasking learning achieves better performance than statistical machine translation in 88% of cases.

Improving Machine Translation of Arabic Dialects through Multi-Task Learning

In this paper, we present the results of an empirical evaluation investigating how recommendation algorithms are affected by popularity bias. Popularity bias makes more popular items to be recommended more frequently than less popular ones, thus it is one of the most relevant issues that limits the fairness of recommender systems. In particular, we define an experimental protocol based on two state-of-theart datasets containing users’ preferences on movies and books and three different recommendation paradigms, i.e., collaborative filtering, content-based filtering and graph-based algorithms. In order to evaluate the overall fairness of the recommendations we use well-known metrics such as Catalogue Coverage, Gini Index and Group Average Popularity (ΔGAP). The goal of this paper is: (i) to provide a clear picture of how recommendation techniques are affected by popularity bias; (ii) to trigger further research in the area aimed to introduce methods to mitigate or reduce biases in order to provide fairer recommendations.

Fairness and Popularity Bias in Recommender Systems: an Empirical Evaluation

On the Impact of Social Media Recommendations on Opinion Consensus

Misogynous MEME Recognition: A Preliminary Study

Symbolic knowledge representation and reasoning and deep learning are fundamentally different approaches to artificial intelligence with complementary capabilities. The former are transparent and data-efficient, but they are sensitive to noise and cannot be applied to non-symbolic domains where the data is ambiguous. The latter can learn complex tasks from examples, are robust to noise, but are black boxes; require large amounts of - not necessarily easily obtained - data, and are slow to learn and prone to adversarial examples. Either paradigm excels at certain types of problems where the other paradigm performs poorly. In order to develop stronger AI systems, integrated neuro-symbolic systems that combine artificial neural networks and symbolic reasoning are being sought. In this context, one of the fundamental open problems is how to perform logic-based deductive reasoning over knowledge bases by means of trainable artificial neural networks. In this talk we will present and discuss recent advances made on this topic, and concrete pointers to open research questions.

Neuro-Symbolic Deep Deductive Reasoning

The language system in the human brain

An automated and data-driven rules extraction is crucial for the construction of Fuzzy Inference Systems (FIS). This work presents a method for extracting fuzzy rules based on clustering and association mining through the FP-growth algorithm. First, Self Organizing Maps are used to identify subsets of elements with similar characteristics, separately for each class. Then, the FP-Growth algorithm is applied to each cluster. Elements matching each rule are subdivided in the corresponding classes and only rules showing a predominance of elements belonging to one class are used as fuzzy rules. The method was applied to the construction of a Decision Support System based on FIS for the target prostate biopsy outcome prediction based on six pre-bioptic variables. A dataset containing 1447 patients (824 with positive outcome, 623 with negative outcome) was used. Four and six clusters were identified for the positive and the negative class, respectively. A total of 151 rules were extracted with FP-Growth algorithm and 29 were included in the FIS. The system was able to classify 927 patients out of 1447. On the classi-fied subjects, it reached a sensitivity of 87.5% and a specificity of 58.8%.

Decision Support System for target prostate biopsy outcome prediction: clustering and FP-growth algorithm for fuzzy rules extraction

Cancer patients may struggle with mental wellbeing issues such as distress and depression. As a part of the CAPABLE project, we aim to develop a digital behaviour-change intervention that helps them build positive health habits and improve their wellbeing. The main challenge to the evaluation of the system is the lack of access to real data prior to intervention start. Therefore, first, we created a simulator that mimics patient responses to activity suggestions based on Fogg’s behaviour model. Later we used supervised and reinforcement learning methods to learn the best time of sending the patient prompts. We found that the reinforcement learning methods learn quickly not to over-notify patients and find prompt policies that are more effective in facilitating users in performing target activity than a random notification strategy, but are less effective than adaptive supervised learning method trained to predict patient responsiveness.

From Personalized Timely Notification to Healthy Habit Formation: A Feasibility Study of Reinforcement Learning Approaches on Synthetic Data

We aim to develop a virtual coaching (VC) system which facilitates cancer patients with the formation of positive health habits. The VC should adjust content and delivery of recommendations based on user’s preferences. We conducted a survey that elicited potential VC users’ perceptions about the appropriate context for notification, aspects driving their motivation, and their ability to perform a selected activity in different circumstances. We analyzed 173 survey responses, identified different types of users and formulated suggestions how to adjust application content and notification timing to increase their susceptibility to non-medication health interventions.

Detection Accuracy for Evaluating Compositional Explanations of Units

Downloads

Next from AIxIA 2021

Explainable Artificial Intelligence for Technology Policy Making Using Attribution Networks

Similar lecture

Misogynous MEME Recognition: A Preliminary Study

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES