Artículos científicos

URI permanente para esta colecciónhttp://10.0.96.45:4000/handle/11056/15104

Examinar

Mostrando 1 - 4 de 4

A dialogue of shared discoveries on immigration: a duoethnography of international students in Canada
(Springer, 2019-04-19) Ovie, Glory R.; Barrantes, Lena
International students are believed to contribute signifcantly in education and research as they bring a rich variety of perspectives, experiences, and languages. International students are frequently categorized into one homogenous group; however, this categorization dishonours their complex intersectional diversity and background that provides cultural capital. There is a need to understand the many manifestations of the complex and intersectional diversity in the backgrounds of international students. These students have many diferent reasons to immigrate to developed countries and undertake a rigorous academic program, including pursuing high academic goals, gaining personal knowledge, developing research skills, and widening employment opportunities. Using a duoethnographic dialogical approach, this article focuses on the experiences of two female international PhD students, one from Nigeria and the other from Costa Rica as they embark on a journey of shared self discoveries on their mobility to Canada. Our paper takes a broad perspective on the processes behind mobility coming from diferent cultures and nationalities that meet in Canada. Some of our fndings include the impact of background when transitioning to a new country, the role of refective dialogue when questioning the source of our cultural assumptions and ethical judgments. In addition, we fnd that duoethnography has a strong efect to re-story our own narratives and perspectives. Finally, this dialogue allows us to broaden how we come to understand and extract meaning from our experiences as international students.
Exploring the effects of silent data corruption in distributed deep learning training
(Institute of Electrical and Electronics Engineers (IEEE), 2022-11-02) Rojas, Elvis; Pérez, Diego; Meneses, Esteban
The profound impact of recent developments in artificial intelligence is unquestionable. The applications of deep learning models are everywhere, from advanced natural language processing to highly accurate prediction of extreme weather. Those models have been continuously increasing in complexity, becoming much more powerful than their original versions. In addition, data to train the models is becoming more available as technological infrastructures sense and collect more readings. Consequently, distributed deep learning training is often times necessary to handle intricate models and massive datasets. Running a distributed training strategy on a supercomputer exposes the models to all the considerations of a large-scale machine; reliability is one of them. As supercomputers integrate a colossal number of components, each fabricated on an ever decreasing feature-size, faults are common during execution of programs. A particular type of fault, silent data corruption, is troublesome because the system does not crash and does not immediately give an evident sign of an error. We set out to explore the effects of that type of faults by inspecting how distributed deep learning training strategies cope with bit-flips that affect their internal data structures. We used checkpoint alteration, a technique that permits the study of this phenomenon on different distributed training platforms and with different deep learning frameworks. We evaluated two distributed learning libraries (Distributed Data Parallel and Horovod) and found out Horovod is slightly more resilient to SDCs. However, fault propagation is similar in both cases, and the model is more sensitive to SDCs than the optimizer.
Metodología del Triple balance como herramienta de evaluación
(2023) Atencio Morales, Roy; Lobo Chaves, William; Barrantes Rivera, Jorge
Hoy más que nunca las empresas deben sumar en sus estrategias la responsabilidad social corporativa (RSC), y la responsabilidad social empresarial (RSE), y para ello el Triple Balance es una de las posibles metodologías para evaluar sus efectos y resultados en el tiempo, es simple, económica, fácil de mantener, y llevar datos, así como de valorar los resultados, en la cual cada una de las empresas puede ir agregando variables de causas - efectos, como de los programas y acciones que se realizan para mitigar o cambiar resultados negativos. La metodología del Triple Balance o Triple Botton Line, (TBL siglas en inglés), medidas a menudo y referidas como personas, ganancias y planeta, que si bien no tiene nada de reciente, es sencilla de utilizar, de mantener y poner en práctica con cierta periodicidad, para evaluar resultados de las empresas. Pretende hacer una revisión del concepto y poder sumar de forma positiva a su aplicación y tomarse para las estrategias de modelos de responsabilidad empresarial como formas evaluativas, como uso frecuente de los datos levantados en el proceso interno y externo de la empresa dentro del método operativo
Understanding failures through the lifetime of a top-level supercomputer
(Academic Press Inc., 2021-04-20) Rojas, Elvis; Meneses, Esteban; Jones, Terry; Maxwell, Don
High performance computing systems are required to solve grand challenges in many scientific disciplines. These systems assemble many components to be powerful enough for solving extremely complex problems. An inherent consequence is the intricacy of the interaction of all those components, especially when failures come into the picture. It is crucial to develop an understanding of how these systems fail to design reliable supercomputing platforms in the future. This paper presents the results on studying multi-year failure and workload records of a powerful supercomputer that topped the world rankings. We provide a thorough analysis of the data and characterize the reliability of the system through several dimensions: failure classification, failure-rate modelling, and interplay between failures and workload. The results shed some light on the dynamics of top-level supercomputers and sensitive areas ripe for improvement.

Examinar

Examinando Artículos científicos por browse.metadata.rights "Acceso embargado"