Libros y Capítulos de libros
URI permanente para esta colección
Examinar
Envíos recientes
Ítem Negative sampling for triplet-based loss: improving representation in self-supervised representation learning(Springer, 2024-11-17) Manuel Alejandro Goyo; Hidalgo Barrientos, Mauricio FernandoSignificant strides have been made in artificial neural networks across various fields, necessitating extensive labeled data for effective training. However, the acquisition of such annotated data is both costly and labor-intensive. To address this challenge, Self-Supervised Representation Learning (SSRL) has emerged as a promising solution. One prominent SSRL method, Contrastive Self-Supervised Learning (CSL), enhances feature representations by discerning similarities and differences among samples in the feature space. Yet, accurately identifying dissimilar samples remains a persistent issue, limiting CSL’s effectiveness. In response, an innovative enhancement to CSL is proposed in this paper. Explicit negative sampling strategies using a binary classification algorithm within the feature space are introduced to distinguish between similar and dissimilar features precisely. Additionally, Triplet Loss, originally designed for tasks such as person re-identification and face recognition, is incorporated to further refine feature learning. Experimental evaluations on the CIFAR-10 and SVHN datasets validate the proposed method’s superiority in content-based image retrieval (CBIR) and classification tasks. Significant improvements are demonstrated in metrics such as mean average precision (MAP), accuracy, recall, precision, and F1-score compared to existing techniques. This framework contributes to the advancement of SSRL by enabling scalable neural network training on large datasets with minimal annotation, effectively bridging the gap between supervised and unsupervised learning paradigms.Ítem A study of software architects’ cognitive approaches: kolb’s learning styles inventory in action(Springer, 2024-10-19) Hidalgo Barrientos, Mauricio Fernando; Astudillo, Hernán; Castro, Laura M.The multidisciplinary nature of software architects demands a diverse set of skills, ranging from technical expertise to interpersonal abilities. Within this domain, software architects are responsible for designing systems that adhere to quality standards, meet functional requirements, and align with organizational goals. However, educating or training software architects presents a challenge due to the complexity of their roles and responsibilities. To address this challenge, this paper proposes an approach to understanding the learning styles of software architects using Kolb’s Learning Style Inventory (KLSI). It aims to provide a characterization of their teaching and learning preferences, thus facilitating the design and execution of educational strategies tailored to their specific needs. In conducting our research, we utilized LinkedIn as a platform to distribute the KLSI Test, ultimately gathering a sample comprising 18 Senior and Mid-Senior Software Architects. Through trend analysis of their responses, we consistently observed a discernible pattern. This led us to identify the Deciding Learning Style as the primary approach among the sample of software architects regarding their learning preferences. This finding offers initial insights into the predominant learning style within this profession, providing valuable guidance for educational practitioners and institutions aiming to optimize their training programs for software architects.Ítem Automatons immersed in ocean currents for transformation of biomass into fuel(Springer, 2021) Cañete, Lucio; Córdova, Felisa; Perez de Arce, AndrésWhen observing the Chilean sea from both biotic and mechanical perspective, oceanographers note that the Humboldt Current carries abundant biomass and that the movement of the water itself has the capacity to do work. Taking advantage of these two qualities of the ocean current, this article exposes the sketch of an automated device, the computational simulation when it was conceived and its mathematical model to make efficient the capture of biomass that will be processed, stored and dispatched as biodiesel. Said submerged automaton has a structural configuration that was outlined by cybernetic design resulting in a body that carries out the transformation process by itself, which starts on the side that faces the current with its content of biomass. This raw material is trapped thanks to an intelligent system that informs the reactor about the relative importance of the state variables that its body can control, stimulating those swimming organisms to move in the desired direction. The captured biomass begins its process until it becomes biodiesel by virtue of the mechanical energy provided by the same flow of seawater that affects the reactor. The rear part of the reactor releases both incident water and by-products into the sea without harmful environmental consequences. Some users of this new type of device are armies in time of conflict and merchant marines during algae bloom.Ítem Low carbon energy thecnologies in sustainable energy systems(Elsevier, 2021) Yanine, Fernando; Sanchez-Squella, Antonio; Barrueto, Aldo; Kumar Sahoo, Sarat; Cordova, Felisa; Shah, Dhruv; Parejo, Antonio; Rother, HansLow Carbon Energy Technologies for Sustainable Energy Systems examines, investigates, and integrates current research aimed at operationalizing low carbon technologies within complex transitioning energy economies. Scholarly research has traditionally focused on the technical aspects of exploitation, R&D, operation, infrastructure, and decommissioning, while approaches which can realistically inform their reception and scale-up across real societies and real markets are piecemeal and isolated in separate literatures. Addressing both the technical foundations of each technology together with the sociotechnical ways in which they are spread in markets and societies, this work integrates the technoeconomic assessment of low carbon technologies with direct discussion on legislative and regulatory policies in energy markets. Chapters address issues, such as social acceptance, consumer awareness, environmental valuation systems, and the circular economy, as low carbon technologies expand into energy systems sustainability, sensitivity, and stability. This collective research work is relevant to both researchers and practitioners working in sustainable energy systems. The combination of these features makes it a timely book that is useful and attractive to university students, researchers, academia, and public or private energy policy makers.Ítem Design of smart socket for monitoring of IoT-based intelligent smart energy management system(Springer, 2021) Rao, Challa Krishna; Kumar Sahoo, Sarat; Balamurugan, M.; Yanine, FernandoSmart socket is designed for collecting and sending the data from the various nodes in one field to other fields. Smart socket consists of the Arduino_Uno, XBee, sensors, gateway, computer, USB, and IDE. This works emphasis on design and development of smart socket with wireless capability, this can be used to collect the data from each electrical device by using sensors. An XBee transmitter and receiver node are used for data communication in wireless networks. Real-time data gathered at the central node can be used to prioritize and schedule the appliances. Then, the system analyzes the data to generate control commands to turn the devices attached to the smart socket on or off. This paper presents the operation and functions of smart socket in different sensor network topologies. The results show that the proposed smart socket can correctly read the data from the various nodes and also send it to different nodes of different parameters.