A carregar...
Projeto de investigação
Data Science and Over-Indebtedness: Use of Artificial Intelligence Algorithms in Credit Consumption and Indebtedness Conciliation in Portugal
Financiador
Autores
Publicações
Tourism in Times of Scarcity: Traveling During the Pandemic (Abstract)
Publication . Castagna, Ana; El Fassi, Yasmina; Pinto, Diego Costa; Mattila, AnnA; Vanneschi, Leonardo; NOVA Information Management School (NOVA IMS); Information Management Research Center (MagIC) - NOVA Information Management School
The financial and health limitations imposed by the COVID-19 pandemic require extensive changes in people’s lives, specifically regarding travel. But might consumers use travel as an emotional regulation tool or a reward during pandemic times? The present work sheds light on how COVID-19 activates unexpected travel behaviors and how consumers’ views on traveling shift depending on the scarcity mindset (high vs. low mutability). Paradoxically, results from three studies (N = 889) show that monetary scarcity can increase consumers’ predisposition to travel post-COVID as a way to restore well-being (Study 1). Studies 2 and 3 further investigate this unexpected downstream effect by showing that pandemic effects on travel behavior depend on the way consumers construe scarcity (low vs. high mutability). When consumers frame the pandemic as an external threat with low (vs. high) mutability, they have a more positive attitude to travel during the pandemic and perceive traveling as less risky.
Structural similarity index (SSIM) revisited
Publication . Bakurov, Illya; Buzzelli, Marco; Schettini, Raimondo; Castelli, Mauro; Vanneschi, Leonardo; NOVA Information Management School (NOVA IMS); Information Management Research Center (MagIC) - NOVA Information Management School; Elsevier Science B.V., Amsterdam.
Several contemporaneous image processing and computer vision systems rely upon the full-reference image quality assessment (IQA) measures. The single-scale structural similarity index (SS-SSIM) is one of the most popular measures, and it owes its success to the mathematical simplicity, low computational complexity, and implicit incorporation of Human Visual System’s (HVS) characteristics. In this paper, we revise the original parameters of SSIM and its multi-scale counterpart (MS-SSIM) to increase their correlation with subjective evaluation. More specifically, we exploit the evolutionary computation and the swarm intelligence methods on five popular IQA databases, two of which are dedicated distance-changed databases, to determine the best combination of parameters efficiently. Simultaneously, we explore the effect of different scale selection approaches in the context of SS-SSIM. The experimental results show that with a proper fine-tuning (1) the performance of SS-SSIM and MS-SSIM can be improved, in average terms, by 8% and by 3%, respectively, (2) the SS-SSIM after the so-called standard scale selection achieves similar performance as if applying computationally more expensive state-of-the-art scale selection methods or MS-SSIM; moreover, (3) there is evidence that the parameters learned on a given database can be successfully transferred to other (previously unseen) databases; finally, (4) we propose a new set of reference parameters for SSIM’s variants and provide their interpretation.
How deeply to fine-tune a convolutional neural network
Publication . Kandel, Ibrahem; Castelli, Mauro; NOVA Information Management School (NOVA IMS); Information Management Research Center (MagIC) - NOVA Information Management School; MDPI - Multidisciplinary Digital Publishing Institute
Accurate classification of medical images is of great importance for correct disease diagnosis. The automation of medical image classification is of great necessity because it can provide a second opinion or even a better classification in case of a shortage of experienced medical staff. Convolutional neural networks (CNN) were introduced to improve the image classification domain by eliminating the need to manually select which features to use to classify images. Training CNN from scratch requires very large annotated datasets that are scarce in the medical field. Transfer learning of CNN weights from another large non-medical dataset can help overcome the problem of medical image scarcity. Transfer learning consists of fine-tuning CNN layers to suit the new dataset. The main questions when using transfer learning are how deeply to fine-tune the network and what difference in generalization that will make. In this paper, all of the experiments were done on two histopathology datasets using three state-of-the-art architectures to systematically study the effect of block-wise fine-tuning of CNN. Results show that fine-tuning the entire network is not always the best option; especially for shallow networks, alternatively fine-tuning the top blocks can save both time and computational power and produce more robust classifiers.
Genetic programming for stacked generalization
Publication . Bakurov, Illya; Castelli, Mauro; Gau, Olivier; Fontanella, Francesco; Vanneschi, Leonardo; NOVA Information Management School (NOVA IMS); Information Management Research Center (MagIC) - NOVA Information Management School; Elsevier
In machine learning, ensemble techniques are widely used to improve the performance of both classification and regression systems. They combine the models generated by different learning algorithms, typically trained on different data subsets or with different parameters, to obtain more accurate models. Ensemble strategies range from simple voting rules to more complex and effective stacked approaches. They are based on adopting a meta-learner, i.e. a further learning algorithm, and are trained on the predictions provided by the single algorithms making up the ensemble. The paper aims at exploiting some of the most recent genetic programming advances in the context of stacked generalization. In particular, we investigate how the evolutionary demes despeciation initialization technique, ϵ-lexicase selection, geometric-semantic operators, and semantic stopping criterion, can be effectively used to improve GP-based systems’ performance for stacked generalization (a.k.a. stacking). The experiments, performed on a broad set of synthetic and real-world regression problems, confirm the effectiveness of the proposed approach.
Semantic Segmentation Network Stacking with Genetic Programming
Publication . Bakurov, Illya; Buzzelli, Marco; Schettini, Raimondo; Castelli, Mauro; Vanneschi, Leonardo; Information Management Research Center (MagIC) - NOVA Information Management School; NOVA Information Management School (NOVA IMS); Springer Science Business Media
Semantic segmentation consists of classifying each pixel of an image and constitutes an essential step towards scene recognition and understanding. Deep convolutional encoder–decoder neural networks now constitute state-of-the-art methods in the field of semantic segmentation. The problem of street scenes’ segmentation for automotive applications constitutes an important application field of such networks and introduces a set of imperative exigencies. Since the models need to be executed on self-driving vehicles to make fast decisions in response to a constantly changing environment, they are not only expected to operate reliably but also to process the input images rapidly. In this paper, we explore genetic programming (GP) as a meta-model that combines four different efficiency-oriented networks for the analysis of urban scenes. Notably, we present and examine two approaches. In the first approach, we represent solutions as GP trees that combine networks’ outputs such that each output class’s prediction is obtained through the same meta-model. In the second approach, we propose representing solutions as lists of GP trees, each designed to provide a unique meta-model for a given target class. The main objective is to develop efficient and accurate combination models that could be easily interpreted, therefore allowing gathering some hints on how to improve the existing networks. The experiments performed on the Cityscapes dataset of urban scene images with semantic pixel-wise annotations confirm the effectiveness of the proposed approach. Specifically, our best-performing models improve systems’ generalization ability by approximately 5% compared to traditional ensembles, 30% for the less performing state-of-the-art CNN and show competitive results with respect to state-of-the-art ensembles. Additionally, they are small in size, allow interpretability, and use fewer features due to GP’s automatic feature selection.
Unidades organizacionais
Descrição
Palavras-chave
Contribuidores
Financiadores
Entidade financiadora
Fundação para a Ciência e a Tecnologia
Programa de financiamento
3599-PPCDT
Número da atribuição
DSAIPA/DS/0113/2019