Publicaciones

RL1, Publisher: Advances in Neural Information Processing Systems, Link >

AUTHORS

Julio Hurtado, Alain Raymond, Álvaro Soto

When learning tasks over time, artificial neural networks suffer from a problem known as Catastrophic Forgetting (CF). This happens when the weights of a network are overwritten during the training of a new task causing forgetting of old information. To address this issue, we propose MetA Reusable Knowledge or MARK, a new method that fosters weight reusability instead of overwriting when learning a new task. Specifically, MARK keeps a set of shared weights among tasks. We envision these shared weights as a common Knowledge Base (KB) that is not only used to learn new tasks, but also enriched with new knowledge as the model learns new tasks. Key components behind MARK are two-fold. On the one hand, a metalearning approach provides the key mechanism to incrementally enrich the KB with new knowledge and to foster weight reusability among tasks. On the other hand, a set of trainable masks provides the key mechanism to selectively choose from the KB relevant weights to solve each task. By using MARK, we achieve state of the art results in several popular benchmarks, surpassing the best performing methods in terms of average accuracy by over 10% on the 20-Split-MiniImageNet dataset, while achieving almost zero forgetfulness using 55% of the number of parameters. Furthermore, an ablation study provides evidence that, indeed, MARK is learning reusable knowledge that is selectively used by each task.

25 visualizaciones Ir a la publicación

RL1, Publisher: Computers in biology and medicine, Link >

AUTHORS

José Rodríguez-Palomares, Julio Sotelo, Lydia Dux-Santoy, Pamela Franco, Rodrigo Salas, Sergio Uribe, Domingo Mery, Andrea Guala, Arturo Evangelista

Abstract: Recent advances in medical imaging have confirmed the presence of altered hemodynamics in bicuspid aortic valve (BAV) patients. Therefore, there is a need for new hemodynamic biomarkers to refine disease monitoring and improve patient risk stratification. This research aims to analyze and extract multiple correlation patterns of hemodynamic parameters from 4D Flow MRI data and find which parameters allow an accurate classification between healthy volunteers (HV) and BAV patients with dilated and non-dilated ascending aorta using machine learning. Sixteen hemodynamic parameters were calculated in the ascending aorta (AAo) and aortic arch (AArch) at peak systole from 4D Flow MRI. We used sequential forward selection (SFS) and principal component analysis (PCA) as feature selection algorithms. Then, eleven machine-learning classifiers were implemented to separate HV and BAV patients (non- and dilated ascending aorta). Multiple correlation patterns from hemodynamic parameters were extracted using hierarchical clustering. The linear discriminant analysis and random forest are the best performing classifiers, using five hemodynamic parameters selected with SFS (velocity angle, forward velocity, vorticity, and backward velocity in AAo; and helicity density in AArch) a 96.31 ± 1.76% and 96.00 ± 0.83% accuracy, respectively. Hierarchical clustering revealed three groups of correlated features. According to this analysis, we observed that features selected by SFS have a better performance than those selected by PCA because the five selected parameters were distributed according to 3 different clusters. Based on the proposed method, we concluded that the feature selection method found five potentially hemodynamic biomarkers related to this disease.

22 visualizaciones Ir a la publicación

AUTHORS

Laurence Golborne, Domingo Mery, Susana Figueroa, Daniel Saavedra, Alejandro Kaminetzky

In X-ray testing, the aim is to inspect those inner parts of an object that cannot be detected by the naked eye. Typical applications are the detection of targets like blow holes in casting inspection, cracks in welding inspection, and prohibited objects in baggage inspection. A straightforward solution today is the use of object detection methods based on deep learning models. Nevertheless, this strategy is not effective when the number of available X-ray images for training is low. Unfortunately, the databases in X-ray testing are rather limited. To overcome this problem, we propose a strategy for deep learning training that is performed with a low number of target-free X-ray images with superimposition of many simulated targets. The simulation is based on the Beer–Lambert law that allows to superimpose different layers. Using this method it is very simple to generate training data. The proposed method was used to train known object detection models (e.g. YOLO, RetinaNet, EfficientDet and SSD) in casting inspection, welding inspection and baggage inspection. The learned models were tested on real X-ray images. In our experiments, we show that the proposed solution is simple (the implementation of the training can be done with a few lines of code using open source libraries), effective (average precision was 0.91, 0.60 and 0.88 for casting, welding and baggage inspection respectively), and fast (training was done in a couple of hours, and testing can be performed in 11ms per image). We believe that this strategy makes a contribution to the implementation of practical solutions to the problem of target detection in X-ray testing.

25 visualizaciones Ir a la publicación

RL1, Publisher: , Link >

AUTHORS

Júlio Barreto Guedes da Costa, Leandro Balby Marinho, Rodrygo LT Santos, Denis Parra

Embeddings are core components of modern model-based Collaborative Filtering (CF) methods, such as Matrix Factorization (MF) and Deep Learning variations. In essence, embeddings are mappings of the original sparse representation of categorical features (eg, user and items) to dense low-dimensional representations. A well-known limitation of such methods is that the learned embeddings are opaque and hard to explain to the users. On the other hand, a key feature of simpler KNN-based CF models (aka user/item-based CF) is that they naturally yield similarity-based explanations, ie, similar users/items as evidence to support model recommendations. Unlike related works that try to attribute explicit meaning (via metadata) to the learned embeddings, in this paper, we propose to equip the learned embeddings of MF with meaningful similarity-based explanations. First, we show that the learned user/item …


18 visualizaciones Ir a la publicación

AUTHORS

Felipe Bravo, Juglar Diaz, Barbara Poblete

Abstract
The popularity of mobile devices with GPS capabilities, along with the worldwide adoption of social media, have created a rich source of text data combined with spatio-temporal information. Text data collected from location-based social networks can be used to gain space–time insights into human behavior and provide a view of time and space from the social media lens. From a data modeling perspective, text, time, and space have different scales and representation approaches; hence, it is not trivial to jointly represent them in a unified model. Existing approaches do not capture the sequential structure present in texts or the patterns that drive how text is generated considering the spatio-temporal context at different levels of granularity. In this work, we present a neural language model architecture that allows us to represent time and space as context for text generation at different granularities. We define the task of modeling text, timestamps, and geo-coordinates as a spatio-temporal conditioned language model task. This task definition allows us to employ the same evaluation methodology used in language modeling, which is a traditional natural language processing task that considers the sequential structure of texts. We conduct experiments over two datasets collected from location-based social networks, Twitter and Foursquare. Our experimental results show that each dataset has particular patterns for language generation under spatio-temporal conditions at different granularities. In addition, we present qualitative analyses to show how the proposed model can be used to characterize urban places.

18 visualizaciones Ir a la publicación

RL3, Publisher: Complexity, Link >

AUTHORS

Tomas Veloz, Marta Lenartowicz, Francis Heylighen, Pedro Maldonado, Shima Beigi, Evo Bussseniers, Alejandro Bassi

Abstract

Reaction network is a promising framework for representing complex systems of diverse and even interdisciplinary types. In this approach, complex systems appear as self-maintaining structures emerging from a multitude of interactions, similar to proposed scenarios for the origin of life out of autocatalytic networks. The formalism of chemical organization theory (COT) mathematically specifies under which conditions a reaction network is stable enough to be observed as a whole complex system. Such conditions specify the notion of organization, crucial in COT. In this paper, we show that the structure and operation of organizations can be advanced towards a formal framework of resilience in complex systems. That is, we show that there exist three fundamental types of change (state, process, and structural) defined for reaction networks, and that these perturbations not only provide a general representation of perturbations in the context of resilience but also pave the ground to formalize different forms of resilient responses. In particular, we show that decomposing the network’s operational structure into dynamically decoupled modules allows to formalize what is the impact of a perturbation and to what extent any potential compensation to that perturbation will be successful. We illustrate our approach with a toy model of a farm that operates in a sustainable way producing milk, eggs, and/or grains from other resources. With the help of simulations, we analyze the different types of perturbations and responses that the farm can undergo and how that affects its sustainable operation.


18 visualizaciones Ir a la publicación

RL2, Publisher: Dagstuhl Reports, Link >

AUTHORS

Leopoldo Bertossi

Abstract

The presentation turns around the subject of explainable AI. More specifically, we deal with attribution numerical scores that are assigned to features values of an entity under classification, to identify and rank their importance for the obtained classification label. We concentrate on the popular SHAP score [2] that can be applied with black-box and open models. We show that, in contrast to its general #P

AUTHORS

Leopoldo Bertossi

-hardness, it can be computed in polynomial time for classifiers that are based on decomposable and deterministic Boolean decision circuits. This class of classifiers includes decision trees and ordered binary decision diagrams. This result was established in [1]. The presentation illustrates how the proof heavily relies on the connection to SAT-related computational problems.


18 visualizaciones Ir a la publicación

RL1, Publisher: Computer Vision for X-Ray Testing, Link >

AUTHORS

Domingo Mery, Bernardita Morris

Abstract

Given a facial matcher, in explainable face verification, the task is to answer: how relevant are the parts of a probe image to establish the matching with an enrolled image. In many cases, however, the trained models cannot be manipulated and must be treated as "black-boxes". In this paper, we present six different saliency maps that can be used to explain any face verification algorithm with no manipulation inside of the face recognition model. The key idea of the methods is based on how the matching score of the two face images changes when the probe is perturbed. The proposed methods remove and aggregate different parts of the face, and measure contributions of these parts individually and in-collaboration as well. We test and compare our proposed methods in three different scenarios: synthetic images with different qualities and occlusions, real face images with different facial expressions, poses, and occlusions and faces from different demographic groups. In our experiments, five different face verification algorithms are used: ArcFace, Dlib, FaceNet (trained on VGGface2 and Casia-WebFace), and LBP. We conclude that one of the proposed methods achieves saliency maps that are stable and interpretable to humans. In addition, our method, in combination with a new visualization of saliency maps based on contours, shows promising results in comparison with other state-of-the-art art methods. This paper presents good insights into any face verification algorithm, in which it can be clearly appreciated which are the most relevant face areas that an algorithm takes into account to carry out the recognition process.


31 visualizaciones Ir a la publicación