Search

Detecting actions as they occur is essential for applications like video surveillance, autonomous driving, and human-robot interaction. Known as online action detection, this task requires classifying actions in streaming videos, handling background noise, and coping with incomplete actions. Transformer architectures are the current state-of-the-art, yet the potential of recent advancements in computer vision, particularly vision-language models (VLMs), remains largely untapped for this problem, partly due to high computational costs. In this paper, we introduce TOAD: a Text-driven Online Action Detection architecture that supports zero-shot and few-shot learning. TOAD leverages CLIP (Contrastive Language-Image Pretraining) textual embeddings, enabling efficient use of VLMs without significant computational overhead. Our model achieves 82.46% mAP on the THUMOS14 dataset, outperforming existing methods, and sets new baselines for zero-shot and few-shot performance on the THUMOS14 and TVSeries datasets., Comment: Published in Integrated Computer-Aided Engineering

The current biodiversity loss crisis makes animal monitoring a relevant field of study. In light of this, data collected through monitoring can provide essential insights, and information for decision-making aimed at preserving global biodiversity. Despite the importance of such data, there is a notable scarcity of datasets featuring videos of birds, and none of the existing datasets offer detailed annotations of bird behaviors in video format. In response to this gap, our study introduces the first fine-grained video dataset specifically designed for bird behavior detection and species classification. This dataset addresses the need for comprehensive bird video datasets and provides detailed data on bird actions, facilitating the development of deep learning models to recognize these, similar to the advancements made in human action recognition. The proposed dataset comprises 178 videos recorded in Spanish wetlands, capturing 13 different bird species performing 7 distinct behavior classes. In addition, we also present baseline results using state of the art models on two tasks: bird behavior recognition and species classification.

The rapid evolution of digital image manipulation techniques poses significant challenges for content verification, with models such as stable diffusion and mid-journey producing highly realistic, yet synthetic, images that can deceive human perception. This research develops and evaluates convolutional neural networks (CNNs) specifically tailored for the detection of these manipulated images. The study implements a comparative analysis of three progressively complex CNN architectures, assessing their ability to classify and localize manipulations across various facial image modifications. Regularization and optimization techniques were systematically incorporated to improve feature extraction and performance. The results indicate that the proposed models achieve an accuracy of up to 76\% in distinguishing manipulated images from genuine ones, surpassing traditional approaches. This research not only highlights the potential of CNNs in enhancing the robustness of digital media verification tools, but also provides insights into effective architectural adaptations and training strategies for low-computation environments. Future work will build on these findings by extending the architectures to handle more diverse manipulation techniques and integrating multi-modal data for improved detection capabilities.

The sequential execution of actions and their hierarchical structure consisting of different levels of abstraction, provide features that remain unexplored in the task of action recognition. In this study, we present a novel approach to improve action recognition by exploiting the hierarchical organization of actions and by incorporating contextualized textual information, including location and prior actions to reflect the sequential context. To achieve this goal, we introduce a novel transformer architecture tailored for action recognition that utilizes both visual and textual features. Visual features are obtained from RGB and optical flow data, while text embeddings represent contextual information. Furthermore, we define a joint loss function to simultaneously train the model for both coarse and fine-grained action recognition, thereby exploiting the hierarchical nature of actions. To demonstrate the effectiveness of our method, we extend the Toyota Smarthome Untrimmed (TSU) dataset to introduce action hierarchies, introducing the Hierarchical TSU dataset. We also conduct an ablation study to assess the impact of different methods for integrating contextual and hierarchical data on action recognition performance. Results show that the proposed approach outperforms pre-trained SOTA methods when trained with the same hyperparameters. Moreover, they also show a 17.12% improvement in top-1 accuracy over the equivalent fine-grained RGB version when using ground-truth contextual information, and a 5.33% improvement when contextual information is obtained from actual predictions.

Cognitive decline is a natural part of aging, often resulting in reduced cognitive abilities. In some cases, however, this decline is more pronounced, typically due to disorders such as Alzheimer's disease. Early detection of anomalous cognitive decline is crucial, as it can facilitate timely professional intervention. While medical data can help in this detection, it often involves invasive procedures. An alternative approach is to employ non-intrusive techniques such as speech or handwriting analysis, which do not necessarily affect daily activities. This survey reviews the most relevant methodologies that use deep learning techniques to automate the cognitive decline estimation task, including audio, text, and visual processing. We discuss the key features and advantages of each modality and methodology, including state-of-the-art approaches like Transformer architecture and foundation models. In addition, we present works that integrate different modalities to develop multimodal models. We also highlight the most significant datasets and the quantitative results from studies using these resources. From this review, several conclusions emerge. In most cases, the textual modality achieves the best results and is the most relevant for detecting cognitive decline. Moreover, combining various approaches from individual modalities into a multimodal model consistently enhances performance across nearly all scenarios.

In this work we rigorously establish mathematical models to obtain the capital valuation adjustment (KVA) as part of the total valuation adjustments (XVAs). For this purpose, we use a semi-replication strategy based on market theory. We formulate single-factor models in terms of expectations and PDEs. For PDEs formulation, we rigorously obtain the existence and uniqueness of the solution, as well as some regularity and qualitative properties of the solution. Moreover, appropriate numerical methods are proposed for solving the corresponding PDEs. Finally, some examples show the numerical results for call and put European options and the corresponding XVA that includes the KVA.

In this article, we propose novel boundary treatment algorithms to avoid order reduction when implicit-explicit Runge-Kutta time discretization is used for solving convection-diffusion-reaction problems with time-dependent Di\-richlet boundary conditions. We consider Cartesian meshes and PDEs with stiff terms coming from the diffusive parts of the PDE. The algorithms treat boundary values at the implicit-explicit internal stages in the same way as the interior points. The boundary treatment strategy is designed to work with multidimensional problems with possible nonlinear advection and source terms. The proposed methods recover the designed order of convergence by numerical verification. For the spatial discretization, in this work, we consider Local Discontinuous Galerkin methods, although the developed boundary treatment algorithms can operate with other discretization schemes in space, such as Finite Differences, Finite Elements or Finite Volumes.

In this article we present a novel and general methodology for building second order finite volume implicit-explicit (IMEX) numerical schemes for solving two dimensional financial parabolic PDEs with mixed derivatives. In particular, applications to basket and Heston models are presented. The obtained numerical schemes have excellent properties and are able to overcome the well-documented difficulties related with numerical approximations in the financial literature. The methods achieve true second order convergence with non-regular initial conditions. Besides, the IMEX time integrator allows to overcome the tiny time-step induced by the diffusive term in the explicit schemes, also providing very accurate and non-oscillatory approximations of the Greeks. Finally, in order to assess all the aforementioned good properties of the developed numerical schemes, we compute extremely accurate semi-analytic solutions using multi-dimensional Fourier cosine expansions. A novel technique to truncate the Fourier series for basket options is presented and it is efficiently implemented using multi-GPUs., Comment: arXiv admin note: substantial text overlap with arXiv:2409.01131

The United Nations 2030 Agenda for Sustainable Development outlines 17 goals to address global challenges. However, progress has been slower than expected and, consequently, there is a need to investigate the reasons behind this fact. In this study, we used a novel data-driven methodology to analyze data from 107 countries (2000$-$2022) using unsupervised machine learning techniques. Our analysis reveals strong positive and negative correlations between certain SDGs. The findings show that progress toward the SDGs is heavily influenced by geographical, cultural and socioeconomic factors, with no country on track to achieve all goals by 2030. This highlights the need for a region specific, systemic approach to sustainable development that acknowledges the complex interdependencies of the goals and the diverse capacities of nations. Our approach provides a robust framework for developing efficient and data-informed strategies, to promote cooperative and targeted initiatives for sustainable progress.

We present a novel and general methodology for building second-order finite volume implicit-explicit Runge-Kutta numerical schemes for solving two-dimensional financial parabolic PDEs with mixed derivatives. The methods achieve second-order convergence even in the presence of non-regular initial conditions. The IMEX time integrator allows to overcome the tiny time-step induced by the diffusive term in the explicit schemes, also providing accurate and non-oscillatory approximations of the Greeks.

The goal of this paper is to develop 2nd order Implicit-Explicit Runge-Kutta (IMEX-RK) finite volume (FV) schemes for solving 1d parabolic PDEs for option pricing, with possible nonlinearities in the source and advection terms. The spatial semi-discretization of the advection is carried out by combining finite volume methods with 2nd order state reconstructions; while the diffusive terms are discretized using second-order finite differences. The time integration is performed by means of IMEX-RK time integrators: the advection is treated explicitly, and the diffusion, implicitly. The obtained numerical schemes have several advantages: they are computationally very efficient, thanks to the implicit discretization of the diffusion in the IMEX-RK time integrators, which allows to overcome the strict time step restriction; they yield second-order accuracy for even nonlinear problems and with non-regular initial conditions; and they can be extended to higher order.

The goal of this article is to make automatic data assimilation for a landslide tsunami model, given by the coupling between a non-hydrostatic multi-layer shallow-water and a Savage-Hutter granular landslide model for submarine avalanches. The coupled model is discretized using a positivity-preserving second-order path-conservative finite volume scheme. The data assimilation problem is posed in a global optimization framework and we develop and compare parallel metaheuristic stochastic global optimization algorithms, more precisely multi-path versions of the Simulated Annealing algorithm, with hybrid global optimization algorithms based on hybridizing Simulated Annealing with gradient local searchers, like L-BGFS-B.

In a recent work, assuming a Beer-Lambert optical absorption and a Gaussian laser time profile, the exact solutions for a 1D-photoacoustic(PA)-boundary value problem predict a null pressure for optically strong absorbent materials. To overcome this, a heuristic correction was introduced by assuming that heat flux travels a characteristic length during the duration of the laser pulse\cite{Ruiz-Veloz2021} $\tau_p$. In this work, we obtained exact solutions in the frequency domain for a 1D-boundary-value-problem for the Dual-Phase-Lag (DPL) heat equation coupled with a 1D PA-boundary-value-problem via the wave-equation. Temperature and pressure solutions were studied by assuming that the sample and its surroundings have a similar characteristic thermal lag response time $\tau_{_T}$, which was assumed to be a free parameter that can be adjusted to reproduce experimental results. Solutions for temperature and pressure were obtained for a three-layer 1D system. It was found that for $\tau_{_T}< \tau_{p}$, the DPL temperature has a similar thermal profile of the Fourier heat equation, however, when $\tau_{_T}\ge \tau_{p}$ this profile is very different from the Fourier case. Additionally, via a numerical Fourier transform the wave-like behavior of DPL temperature is explored, and it was found that as $\tau_{_T}$ increases, thermal wave amplitude is increasingly attenuated. Exact solutions for pressure were compared with experimental signals, showing a close resemblance between both data sets, particularly in the time domain, for an appropriated value of $\tau_{_T}$; the transference function was also calculated, which allowed us to find the maximum response in frequency for the considered experimental setup., Comment: 32 pages, 11 figures

Cognitive decline is a natural process that occurs as individuals age. Early diagnosis of anomalous decline is crucial for initiating professional treatment that can enhance the quality of life of those affected. To address this issue, we propose a multimodal model capable of predicting Mild Cognitive Impairment and cognitive scores. The TAUKADIAL dataset is used to conduct the evaluation, which comprises audio recordings of clinical interviews. The proposed model demonstrates the ability to transcribe and differentiate between languages used in the interviews. Subsequently, the model extracts audio and text features, combining them into a multimodal architecture to achieve robust and generalized results. Our approach involves in-depth research to implement various features obtained from the proposed modalities., Comment: GitHub repository: https://github.com/davidorp/taukadial

Generating human-human motion interactions conditioned on textual descriptions is a very useful application in many areas such as robotics, gaming, animation, and the metaverse. Alongside this utility also comes a great difficulty in modeling the highly dimensional inter-personal dynamics. In addition, properly capturing the intra-personal diversity of interactions has a lot of challenges. Current methods generate interactions with limited diversity of intra-person dynamics due to the limitations of the available datasets and conditioning strategies. For this, we introduce in2IN, a novel diffusion model for human-human motion generation which is conditioned not only on the textual description of the overall interaction but also on the individual descriptions of the actions performed by each person involved in the interaction. To train this model, we use a large language model to extend the InterHuman dataset with individual descriptions. As a result, in2IN achieves state-of-the-art performance in the InterHuman dataset. Furthermore, in order to increase the intra-personal diversity on the existing interaction datasets, we propose DualMDM, a model composition technique that combines the motions generated with in2IN and the motions generated by a single-person motion prior pre-trained on HumanML3D. As a result, DualMDM generates motions with higher individual diversity and improves control over the intra-person dynamics while maintaining inter-personal coherence., Comment: Project page: https://pabloruizponce.github.io/in2IN/

As quantum computers advance, the complexity of the software they can execute increases as well. To ensure this software is efficient, maintainable, reusable, and cost-effective -key qualities of any industry-grade software-mature software engineering practices must be applied throughout its design, development, and operation. However, the significant differences between classical and quantum software make it challenging to directly apply classical software engineering methods to quantum systems. This challenge has led to the emergence of Quantum Software Engineering as a distinct field within the broader software engineering landscape. In this work, a group of active researchers analyse in depth the current state of quantum software engineering research. From this analysis, the key areas of quantum software engineering are identified and explored in order to determine the most relevant open challenges that should be addressed in the next years. These challenges help identify necessary breakthroughs and future research directions for advancing Quantum Software Engineering., Comment: Extended version of the previous FSE paper for the ACM TOSEM Special Issue

This study shows the feasibility of an eHealth solution for tackling eating habits and physical activity in the adolescent population. The participants were children from 11 to 15 years old. An intervention was carried out on 139 students in the intervention group and 91 students in the control group, in two schools during 14 weeks. The intervention group had access to the web through a user account and a password. They were able to create friendship relationships, post comments, give likes and interact with other users, as well as receive notifications and information about nutrition and physical activity on a daily basis and get (virtual) rewards for improving their habits. The control group did not have access to any of these features. The homogeneity of the samples in terms of gender, age, body mass index and initial health-related habits was demonstrated. Pre- and post-measurements were collected through self-reports on the application website. After applying multivariate analysis of variance, a significant alteration in the age-adjusted body mass index percentile was observed in the intervention group versus the control group, as well as in the PAQ-A score and the KIDMED score. It can be concluded that eHealth interventions can help to obtain healthy habits. More research is needed to examine the effectiveness in achieving adherence to these new habits.

Alcohol Use Disorder (AUD) is a major concern for public health organizations worldwide, especially as regards the adolescent population. The consumption of alcohol in adolescents is known to be influenced by seeing friends and even parents drinking alcohol. Building on this fact, a number of studies into alcohol consumption among adolescents have made use of Social Network Analysis (SNA) techniques to study the different social networks (peers, friends, family, etc.) with whom the adolescent is involved. These kinds of studies need an initial phase of data gathering by means of questionnaires and a subsequent analysis phase using the SNA techniques. The process involves a number of manual data handling stages that are time consuming and error-prone. The use of knowledge engineering techniques (including the construction of a domain ontology) to represent the information, allows the automation of all the activities, from the initial data collection to the results of the SNA study. This paper shows how a knowledge model is constructed, and compares the results obtained using the traditional method with this, fully automated model, detailing the main advantages of the latter. In the case of the SNA analysis, the validity of the results obtained with the knowledge engineering approach are compared to those obtained manually using the UCINET, Cytoscape, Pajek and Gephi to test the accuracy of the knowledge model.

The spread of the Internet of Things (IoT) is demanding new, powerful architectures for handling the huge amounts of data produced by the IoT devices. In many scenarios, many existing isolated solutions applied to IoT devices use a set of rules to detect, report and mitigate malware activities or threats. This paper describes a development environment that allows the programming and debugging of such rule-based multi-agent solutions. The solution consists of the integration of a rule engine into the agent, the use of a specialized, wrapping agent class with a graphical user interface for programming and testing purposes, and a mechanism for the incremental composition of behaviors. Finally, a set of examples and a comparative study were accomplished to test the suitability and validity of the approach. The JADE multi-agent middleware has been used for the practical implementation of the approach., Comment: arXiv admin note: substantial text overlap with arXiv:2402.09499

The increasing number of connected devices and the complexity of Internet of Things (IoT) ecosystems are demanding new architectures for managing and securing these networked environments. Intrusion Detection Systems (IDS) are security solutions that help to detect and mitigate the threats that IoT systems face, but there is a need for new IDS strategies and architectures. This paper describes a development environment that allows the programming and debugging of distributed, rule-based multi-agent IDS solutions. The proposed solution consists in the integration of a rule engine into the agent, the use of a specialized, wrapping agent class with a graphical user interface for programming and debugging purposes, and a mechanism for the incremental composition of behaviors. A comparative study and an example IDS are used to test and show the suitability and validity of the approach. The JADE multi-agent middleware has been used for the practical implementations.

This work addresses the performance comparison between four clustering techniques with the objective of achieving strong hybrid models in supervised learning tasks. A real dataset from a bio-climatic house named Sotavento placed on experimental wind farm and located in Xermade (Lugo) in Galicia (Spain) has been collected. Authors have chosen the thermal solar generation system in order to study how works applying several cluster methods followed by a regression technique to predict the output temperature of the system. With the objective of defining the quality of each clustering method two possible solutions have been implemented. The first one is based on three unsupervised learning metrics (Silhouette, Calinski-Harabasz and Davies-Bouldin) while the second one, employs the most common error measurements for a regression algorithm such as Multi Layer Perceptron.

This study proposes a method based on fully convolutional neural networks (FCNs) to identify migratory birds from their songs, with the objective of recognizing which birds pass through certain areas and at what time. To determine the best FCN architecture, extensive experimentation was conducted through a grid search, exploring the optimal depth, width, and activation function of the network. The results showed that the optimal number of filters is 400 in the widest layer, with 4 convolutional blocks with maxpooling and an adaptive activation function. The proposed FCN offers a significant advantage over other techniques, as it can recognize the sound of a bird in audio of any length with an accuracy greater than 85%. Furthermore, due to its architecture, the network can detect more than one species from audio and can carry out near-real-time sound recognition. Additionally, the proposed method is lightweight, making it ideal for deployment and use in IoT devices. The study also presents a comparative analysis of the proposed method against other techniques, demonstrating an improvement of over 67% in the best-case scenario. These findings contribute to advancing the field of bird sound recognition and provide valuable insights into the practical application of FCNs in real-world scenarios.

Background and objective: Diabetes is a chronic pathology which is affecting more and more people over the years. It gives rise to a large number of deaths each year. Furthermore, many people living with the disease do not realize the seriousness of their health status early enough. Late diagnosis brings about numerous health problems and a large number of deaths each year so the development of methods for the early diagnosis of this pathology is essential. Methods: In this paper, a pipeline based on deep learning techniques is proposed to predict diabetic people. It includes data augmentation using a variational autoencoder (VAE), feature augmentation using an sparse autoencoder (SAE) and a convolutional neural network for classification. Pima Indians Diabetes Database, which takes into account information on the patients such as the number of pregnancies, glucose or insulin level, blood pressure or age, has been evaluated. Results: A 92.31% of accuracy was obtained when CNN classifier is trained jointly the SAE for featuring augmentation over a well balanced dataset. This means an increment of 3.17% of accuracy with respect the state-of-the-art. Conclusions: Using a full deep learning pipeline for data preprocessing and classification has demonstrate to be very promising in the diabetes detection field outperforming the state-of-the-art proposals.

In this work, a sentiment analysis method that is capable of accepting audio of any length, without being fixed a priori, is proposed. Mel spectrogram and Mel Frequency Cepstral Coefficients are used as audio description methods and a Fully Convolutional Neural Network architecture is proposed as a classifier. The results have been validated using three well known datasets: EMODB, RAVDESS, and TESS. The results obtained were promising, outperforming the state-of-the-art methods. Also, thanks to the fact that the proposed method admits audios of any size, it allows a sentiment analysis to be made in near real time, which is very interesting for a wide range of fields such as call centers, medical consultations, or financial brokers.

The aim of this paper was the detection of pathologies through respiratory sounds. The ICBHI (International Conference on Biomedical and Health Informatics) Benchmark was used. This dataset is composed of 920 sounds of which 810 are of chronic diseases, 75 of non-chronic diseases and only 35 of healthy individuals. As more than 88% of the samples of the dataset are from the same class (Chronic), the use of a Variational Convolutional Autoencoder was proposed to generate new labeled data and other well known oversampling techniques after determining that the dataset classes are unbalanced. Once the preprocessing step was carried out, a Convolutional Neural Network (CNN) was used to classify the respiratory sounds into healthy, chronic, and non-chronic disease. In addition, we carried out a more challenging classification trying to distinguish between the different types of pathologies or healthy: URTI, COPD, Bronchiectasis, Pneumonia, and Bronchiolitis. We achieved results up to 0.993 F-Score in the three-label classification and 0.990 F-Score in the more challenging six-class classification.

The use of social network theory and methods of analysis have been applied to different domains in recent years, including public health. The complete procedure for carrying out a social network analysis (SNA) is a time-consuming task that entails a series of steps in which the expert in social network analysis could make mistakes. This research presents a multi-domain knowledge model capable of automatically gathering data and carrying out different social network analyses in different domains, without errors and obtaining the same conclusions that an expert in SNA would obtain. The model is represented in an ontology called OntoSNAQA, which is made up of classes, properties and rules representing the domains of People, Questionnaires and Social Network Analysis. Besides the ontology itself, different rules are represented by SWRL and SPARQL queries. A Knowledge Based System was created using OntoSNAQA and applied to a real case study in order to show the advantages of the approach. Finally, the results of an SNA analysis obtained through the model were compared to those obtained from some of the most widely used SNA applications: UCINET, Pajek, Cytoscape and Gephi, to test and confirm the validity of the model.