Chronological and Hierarchical Reasoning Over Naturally Occurring Schemas (CHRONOS) is a system that combines language model-based natural language processing with symbolic knowledge representations to analyze and make predictions about newsworthy events. CHRONOS consists of an event-centric information extraction pipeline and a complex event schema instantiation and prediction system. Resulting predictions are detailed with arguments, event types from Wikidata, schema-based justifications, and source document provenance. We evaluate our system by its ability to capture the structure of unseen events described in news articles and make plausible predictions as judged by human annotators.

Monocular depth estimation is gaining attention in computer vision for autonomous driving due to its cost-effectiveness and versatility. Recent works have used self-supervised learning to tackle this problem. However, they overlook crucial challenges for real-world deployment, such as data privacy, network consumption, computational cost, and unstable connectivity. Recent studies have shown the potential benefits of federated learning in addressing these challenges. Thus, we introduce FedSCDepth, a novel method that combines federated learning and deep self-supervision to efficiently learn monocular depth estimators with performance comparable to the state-of-the-art. Our evaluation using the KITTI dataset’s Eigen’s Split reveals that FedSCDepth achieves high efficacy, with a test loss below 0.13. It also demonstrates impressive efficiency, requiring only an average of 1.5k training steps and up to 0.415 GB of weight data transfer per autonomous vehicle in each round.

Recent investigations in computer vision for autonomous vehicles have focused on depth estimation from images, owing to its cost-efficiency and adaptability. Monocular depth estimation, using a single camera, is notable for its adaptability compared to binocular techniques that require two fixed cameras. Sophisticated methodologies employ self-supervised deep neural networks, while latest research proposes the use of federated learning to tackle crucial challenges for autonomous driving, such as data privacy, network usage, computation distribution, and connectivity robustness. Nevertheless, continual learning presents an additional challenge for both centralized and federated training, as updating models with novel datasets may induce the forgetting of previously acquired samples, thereby diminishing the model’s versatility. Meanwhile, recent research indicates that experience replay strategies can be used to alleviate forgetting in autonomous driving use cases. We introduce ERFedSCDepth, an innovative approach amalgamating experience replay, federated learning, and deep self-supervision to allow the training of monocular depth estimators with high effectiveness and efficiency. Assessment using KITTI and DDAD datasets shows the efficacy of our approach, achieving enhanced continual learning performance over state-of-the-art baseline at standard depth metrics.

Increasing the availability of applications is a way to improve service delivery to final consumers. This paper proposes an architecture based on fog computing and attends intelligent transportation system applications with stricter latency requirements. We also present an algorithm based on autoregressive neural networks and adaptive elasticity thresholds to guarantee applications' availability. The proposed algorithm aims to decrease the number of lost requests coming from the terminal layer. Based on workload, this algorithm analyzes the server resources in a fog-based solution and indicates the best time to increase or decrease the infrastructure. This paper also presents an experimental assessment of different scenarios to manage applications' elasticity using several techniques and policies. Finally, we compare the proposed algorithm with state-of-the-art, and its results showed a reduction of 17.8% in the number of requests lost than the others.

Abstract Machine learning (ML) has already fundamentally changed several businesses. More recently, it has also been profoundly impacting the computational science and engineering domains, like geoscience, climate science, and health science. In these domains, users need to perform comprehensive data analyses combining scientific data and ML models to provide for critical requirements, such as reproducibility, model explainability, and experiment data understanding. However, scientific ML is multidisciplinary, heterogeneous, and affected by the physical constraints of the domain, making such analyses even more challenging. In this work, we leverage workflow provenance techniques to build a holistic view to support the lifecycle of scientific ML. We contribute with (i) characterization of the lifecycle and taxonomy for data analyses; (ii) design decisions to build this view, with a W3C PROV compliant data representation and a reference system architecture; and (iii) lessons learned after an evaluation in an Oil & Gas case using an HPC cluster with 393 nodes and 946 GPUs. The experiments show that the decisions enable queries that integrate domain semantics with ML models while keeping low overhead (<1%), high scalability, and an order of magnitude of query acceleration under certain workloads against without our representation.

Using Edge Computing approaches is a way to improve the Intelligent Transportation Systems (ITS) applications in smart cities. This article proposes a case study of an ITS system with a Mobile Edge Computing (MEC) architecture using the platform for time-sensitive ITS applications. The platform aims to support ITS applications with strict requirements for latency and requests loss. We evaluated an ITS application in three different scenarios: in the cloud using a 4G connection, in the cloud using broadband, and in MEC. We compare scenarios using quantitative metrics such as requests per second, average response time, and request loss rate. The experimental results indicate that MEC has a lower loss of requests and it can work with a greater number of requests per second, which makes it a good solution for ITS applications.

Microservices Architecture has been adopted by several companies to develop applications and replace monolithic ones. Several works point out that this approach supports the design of maintainable software systems. However, none of them presents a quantitative empirical study on the extent of the maintenance support in a real-world application. This work assesses how Microservices Architecture supports software maintenance through an empirical quantitative study of a scientific application built from scratch. We collected data from January 2016 (the beginning of the project) to December 2019, and analyzed 19 microservices, 34 repositories, and 15,408 commits. Then, we present the lessons learned during the project that allowed reaching the assessment results. Our findings may assist practitioners in making architectural decisions and pointing out research opportunities for academics.

Computational Science and Engineering (CSE) projects are typically developed by multidisciplinary teams. Despite being part of the same project, each team manages its own workflows, using specific execution environments and data processing tools. Analyzing the data processed by all workflows globally is a core task in a CSE project. However, this analysis is hard because the data generated by these workflows are not integrated. In addition, since these workflows may take a long time to execute, data analysis needs to be done at runtime to reduce cost and time of the CSE project. A typical solution in scientific data analysis is to capture and relate the data in a provenance database while the workflows run, thus allowing for data analysis at runtime. However, the main problem is that such data capture competes with the running workflows, adding significant overhead to their execution. To mitigate this problem, we introduce in this paper a system called ProvLake, which adopts design principles for providing efficient distributed data capture from the workflows. While capturing the data, ProvLake logically integrates and ingests them into a provenance database ready for analyses at runtime. We validated ProvLake in a real use case in the O&G industry encompassing four workflows that process 5 TB datasets for a deep learning classifier. Compared with Komadu, the closest solution that meets our goals, our approach enables runtime multiworkflow data analysis with much smaller overhead, such as 0.1%.

Intelligent Transportation System's main objective is to provide sustainable and optimized means of transportation for citizens of urban centers. Identifying the travel modes used by citizens, in real- time, allows these systems to better adapt transportation infrastructure and services according to user needs and possibilities of interaction. Many works have explored the use of machine learning algorithms and ensemble methods, combined with general and domain specific feature extraction techniques, for this task. More recently, several works evaluated the use of deep learning algorithms while none of them has explored the use of Recurrent Neural Networks (RNNs) in conjunction with domain feature extraction to enable the development of flexible and lightweight travel mode detection solutions based on multiple smartphone sensor readings. In this paper, we propose a deep RNN architecture for building online travel mode detection models using Long-Short Term Memory (LSTM) cells and evaluate its performance using real mobility data collected with a wide variety of sensors. The experiments show that the proposed architecture allows the generation of models with high accuracy and lower memory consumption and computation cost than state-of-the-art supervised machine learning (ML) approaches.

Identifying the travel modes people use on their daily commute, and the purpose of their trips, contributes to a better understanding of their mobility patterns. Therefore, the detection of the travel modes and the prediction of trips, purposes from smartphone sensors data emerged as two research problems in recent years. Most previous works have tackled these problems in separate studies, leading to an unnecessary redundancy with respect to the preprocessing steps taken. In this paper, we propose the use of a single preprocessing algorithm for both problems, using location traces obtained through smartphone sensors. The proposed technique applies multiple preprocessing steps to extract relevant features for both classification tasks and uses automated machine learning methods to identify the best classifiers and its hyperparameter configurations for each one of them. During evaluation experiments conducted with real smartphones data, a maximum accuracy of 88% for travel mode detection and 81% for trip purpose prediction were reached using our technique.

Context-aware applications in intelligent transportation systems have a growing need for travel mode detection systems. However, few applications allow real-time travel mode detection through the use of smartphones. In this paper, we propose a real-time travel mode detection application based on GPS traces using a data mining technique through which these traces are preprocessed, grouped in motion segments and classified by supervised machine learning algorithms. An application prototype was implemented on the Android platform, used by smartphones, for movement data collection and user travel mode detection using the WEKA API in Java. Finally, to evaluate the performance of the application in a real environment, field tests were carried out with dozens of volunteers in the metropolitan area of Rio de Janeiro. Therefore, 1338 travel mode inferences were obtained by four machine learning techniques and the results were evaluated and compared by the indicators of the confusion matrix. Thus, through the performance evaluation carried out, it was possible to verify that the proposed application is useful for real-time travel mode detection in urban centers.