Search

In Machine Learning scenarios, privacy is a crucial concern when models have to be trained with private data coming from users of a service, such as a recommender system, a location-based mobile service, a mobile phone text messaging service providing next word prediction, or a face image classification system. The main issue is that, often, data are collected, transferred, and processed by third parties. These transactions violate new regulations, such as GDPR. Furthermore, users usually are not willing to share private data such as their visited locations, the text messages they wrote, or the photo they took with a third party. On the other hand, users appreciate services that work based on their behaviors and preferences. In order to address these issues, Federated Learning (FL) has been recently proposed as a means to build ML models based on private datasets distributed over a large number of clients, while preventing data leakage. A federation of users is asked to train a same global model on their private data, while a central coordinating server receives locally computed updates by clients and aggregate them to obtain a better global model, without the need to use clients' actual data. In this work, we extend the FL approach by pushing forward the state-of-the-art approaches in the aggregation step of FL, which we deem crucial for building a high-quality global model. Specifically, we propose an approach that takes into account a suite of client-specific criteria that constitute the basis for assigning a score to each client based on a priority of criteria defined by the service provider. Extensive experiments on two publicly available datasets indicate the merits of the proposed approach compared to standard FL baseline. [ABSTRACT FROM AUTHOR]

AI techniques and systems are pervasive to the media and entertainment industry, with application ranging from chatbots and characters to games and virtual environments. A common feature characterising these applications is given by the intent to introduce a narrative element in the user experience, often conveyed through some type of performance. In this paper, we analyse the contribution of AI techniques in the design and realization of a dramatic performance, an interactive system participated by human performers and audiences through some type of enactment. Drawing on real applications developed for innovative performances, we propose an architectural model that forms the technical platform of the system, and discuss how it can be deployed using Artificial Intelligence techniques with reference to real, experimental applications created in the last two decades. [ABSTRACT FROM AUTHOR]

Recurrent Neural Networks (RNNs) represent a natural paradigm for modeling sequential data like text written in natural language. In fact, RNNs and their variations have long been the architecture of choice in many applications, however in practice they require the use of labored architectures (such as gating mechanisms) and computationally heavy training processes. In this paper we address the question of whether it is possible to generate sentence embeddings via completely untrained recurrent dynamics, on top of which to apply a simple learning algorithm for text classification. This would allow to obtain extremely efficient models in terms of training time. Our work investigates the extent to which this approach can be used, by analyzing the results on different tasks. Finally, we show that, within certain limits, it is possible to build extremely efficient models for text classification that remain competitive in accuracy with reference models in the state-of-the-art. [ABSTRACT FROM AUTHOR]

In this work we present PRONOM , a theorem prover and countermodel generator for non-normal modal logics. PRONOM implements some labelled sequent calculi recently introduced for the basic system E and its extensions with axioms M, N, and C based on bi-neighbourhood semantics. PRONOM is inspired by the methodology of leanTAP and is implemented in Prolog. When a modal formula is valid, then PRONOM computes a proof (a closed tree) in the labelled calculi having a sequent with an empty left-hand side and containing only that formula on the right-hand side as a root, otherwise PRONOM is able to extract a model falsifying it from an open, saturated branch. The paper shows some experimental results, witnessing that the performances of PRONOM are promising. [ABSTRACT FROM AUTHOR]

In recent years, Deep Learning methods have become very popular in classification tasks for Natural Language Processing (NLP); this is mainly due to their ability to reach high performances by relying on very simple input representations, i.e., raw tokens. One of the drawbacks of deep architectures is the large amount of annotated data required for an effective training. Usually, in Machine Learning this problem is mitigated by the usage of semi-supervised methods or, more recently, by using Transfer Learning, in the context of deep architectures. One recent promising method to enable semi-supervised learning in deep architectures has been formalized within Semi-Supervised Generative Adversarial Networks (SS-GANs) in the context of Computer Vision. In this paper, we adopt the SS-GAN framework to enable semi-supervised learning in the context of NLP. We demonstrate how an SS-GAN can boost the performances of simple architectures when operating in expressive low-dimensional embeddings; these are derived by combining the unsupervised approximation of linguistic Reproducing Kernel Hilbert Spaces and the so-called Universal Sentence Encoders. We experimentally evaluate the proposed approach over a semantic classification task, i.e., Question Classification, by considering different sizes of training material and different numbers of target classes. By applying such adversarial schema to a simple Multi-Layer Perceptron, a classifier trained over a subset derived from 1% of the original training material achieves 92% of accuracy. Moreover, when considering a complex classification schema, e.g., involving 50 classes, the proposed method outperforms state-of-the-art alternatives such as BERT. [ABSTRACT FROM AUTHOR]

The Italian Association for Artificial Intelligence (AIxIA) was established in 1988 and has the role of increasing the knowledge of Artificial Intelligence (AI) at several levels, by fostering its teaching in schools and universities, and by promoting both theoretical and applied research in the field through seminars, targeted initiatives and sponsorship of events. The 18th International Conference of the Italian Association for Artificial Intelligence (AIxIA 2019) was held in Rende (CS), Italy, from November 19 to November 22, 2019. [Extracted from the article]