Evaluating Resilience of Deep Learning Models
Deep learning applications have become a valuable tool to solve complex problems in many critical areas. It is important to provide reliability on the outputs of those applications, even if failures occur during execution. In this paper, we present a reliability evaluation of three deep learning mod...
| Formato: | Artículo |
|---|---|
| Idioma: | Inglés |
| Publicado: |
Editorial Tecnológica de Costa Rica (entidad editora)
2020
|
| Materias: | |
| Acceso en línea: |
https://revistas.tec.ac.cr/index.php/tec_marcha/article/view/5071 https://hdl.handle.net/2238/12065 |
| id |
RepoTEC12065 |
|---|---|
| recordtype |
dspace |
| spelling |
RepoTEC120652020-09-25T23:12:52Z Evaluating Resilience of Deep Learning Models Evaluando la Resiliencia de Modelos de Deep Learning Resilience fault tolerance deep learning fault injection Resiliencia tolerancia a fallas deep learning inyección de fallos Deep learning applications have become a valuable tool to solve complex problems in many critical areas. It is important to provide reliability on the outputs of those applications, even if failures occur during execution. In this paper, we present a reliability evaluation of three deep learning models. We use an ImageNet dataset and a homebrew fault injector to make all the tests. The results show there is a difference in failure sensitivity among the models. Also, there are models that despite an increase in the failure rate can keep the resulting error values low. Los modelos de Aprendizaje Profundo se han convertido en una valiosa herramienta para resolver problemas complejos en muchas áreas críticas. Es importante proveer confiabilidad en las salidas de la ejecución de estos modelos, aún si se producen fallos durante la ejecución. En este artículo presentamos la evaluación de la confiabilidad de tres modelos de aprendizaje profundo. Usamos un conjunto de datos de ImageNet y desarrollamos un inyector de fallos para realizar las pruebas. Los resultados muestran que entre los modelos hay una diferencia en la sensibilidad a los fallos. Además, hay modelos que a pesar del incremento en la tasa de fallos pueden mantener bajos los valores de error. 2020-03-27 2020-09-25T23:12:52Z 2020-09-25T23:12:52Z info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion https://revistas.tec.ac.cr/index.php/tec_marcha/article/view/5071 10.18845/tm.v33i5.5071 https://hdl.handle.net/2238/12065 eng https://revistas.tec.ac.cr/index.php/tec_marcha/article/view/5071/4793 application/pdf Editorial Tecnológica de Costa Rica (entidad editora) Tecnología en marcha Journal; 2020: Vol. 33 especial. Contribuciones a la Conferencia 6th Latin America High Performance Computing Conference (CARLA); Pág. 25-30 Revista Tecnología en Marcha; 2020: Vol. 33 especial. Contribuciones a la Conferencia 6th Latin America High Performance Computing Conference (CARLA); Pág. 25-30 2215-3241 0379-3982 |
| institution |
Tecnológico de Costa Rica |
| collection |
Repositorio TEC |
| language |
Inglés |
| topic |
Resilience fault tolerance deep learning fault injection Resiliencia tolerancia a fallas deep learning inyección de fallos |
| spellingShingle |
Resilience fault tolerance deep learning fault injection Resiliencia tolerancia a fallas deep learning inyección de fallos Evaluating Resilience of Deep Learning Models |
| description |
Deep learning applications have become a valuable tool to solve complex problems in many critical areas. It is important to provide reliability on the outputs of those applications, even if failures occur during execution. In this paper, we present a reliability evaluation of three deep learning models. We use an ImageNet dataset and a homebrew fault injector to make all the tests. The results show there is a difference in failure sensitivity among the models. Also, there are models that despite an increase in the failure rate can keep the resulting error values low. |
| format |
Artículo |
| title |
Evaluating Resilience of Deep Learning Models |
| title_short |
Evaluating Resilience of Deep Learning Models |
| title_full |
Evaluating Resilience of Deep Learning Models |
| title_fullStr |
Evaluating Resilience of Deep Learning Models |
| title_full_unstemmed |
Evaluating Resilience of Deep Learning Models |
| title_sort |
evaluating resilience of deep learning models |
| publisher |
Editorial Tecnológica de Costa Rica (entidad editora) |
| publishDate |
2020 |
| url |
https://revistas.tec.ac.cr/index.php/tec_marcha/article/view/5071 https://hdl.handle.net/2238/12065 |
| _version_ |
1796139267295019008 |
| score |
12.2319145 |