How to add new knowledge to already trained deep learning models applied to semantic localization
The capacity of a robot to automatically adapt to new environments is crucial, especially in social robotics. Often, when these robots are deployed in home or office environments, they tend to fail because they lack the ability to adapt to new and continuously changing scenarios. In order to accompl...
Autores Principales: | Cruz, Edmanuel, Rangel, José Carlos, Gomez Donoso, Francisco, Cazorla, Miguel |
---|---|
Formato: | Artículo |
Idioma: | Inglés Inglés |
Publicado: |
2020
|
Materias: | |
Acceso en línea: |
https://link.springer.com/article/10.1007/s10489-019-01517-1 https://ridda2.utp.ac.pa/handle/123456789/9445 https://ridda2.utp.ac.pa/handle/123456789/9445 |
id |
RepoUTP9445 |
---|---|
recordtype |
dspace |
spelling |
RepoUTP94452021-07-06T15:35:10Z How to add new knowledge to already trained deep learning models applied to semantic localization Cruz, Edmanuel Rangel, José Carlos Gomez Donoso, Francisco Cazorla, Miguel Semantic localization Deep learning Retraining strategies Machine learning Semantic localization Deep learning Retraining strategies Machine learning The capacity of a robot to automatically adapt to new environments is crucial, especially in social robotics. Often, when these robots are deployed in home or office environments, they tend to fail because they lack the ability to adapt to new and continuously changing scenarios. In order to accomplish this task, robots must obtain new information from the environment, and then add it to their already learned knowledge. Deep learning techniques are often used to tackle this problem successfully. However, these approaches, complete retraining of the models, which is highly time-consuming. In this work, several strategies are tested to find the best way to include new knowledge in an already learned model in a deep learning pipeline, putting the spotlight on the time spent for this training. We tackle the localization problem in the long term with a deep learning approach and testing several retraining strategies. The results of the experiments are discussed and, finally, the best approach is deployed on a Pepper robot. The capacity of a robot to automatically adapt to new environments is crucial, especially in social robotics. Often, when these robots are deployed in home or office environments, they tend to fail because they lack the ability to adapt to new and continuously changing scenarios. In order to accomplish this task, robots must obtain new information from the environment, and then add it to their already learned knowledge. Deep learning techniques are often used to tackle this problem successfully. However, these approaches, complete retraining of the models, which is highly time-consuming. In this work, several strategies are tested to find the best way to include new knowledge in an already learned model in a deep learning pipeline, putting the spotlight on the time spent for this training. We tackle the localization problem in the long term with a deep learning approach and testing several retraining strategies. The results of the experiments are discussed and, finally, the best approach is deployed on a Pepper robot. 2020-01-02T21:25:41Z 2020-01-02T21:25:41Z 2020-01-02T21:25:41Z 2020-01-02T21:25:41Z 06/19/2019 06/19/2019 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion https://link.springer.com/article/10.1007/s10489-019-01517-1 https://ridda2.utp.ac.pa/handle/123456789/9445 https://ridda2.utp.ac.pa/handle/123456789/9445 eng eng info:eu-repo/semantics/embargoedAccess application/pdf |
institution |
Universidad Tecnológica de Panamá |
collection |
Repositorio UTP – Ridda2 |
language |
Inglés Inglés |
topic |
Semantic localization Deep learning Retraining strategies Machine learning Semantic localization Deep learning Retraining strategies Machine learning |
spellingShingle |
Semantic localization Deep learning Retraining strategies Machine learning Semantic localization Deep learning Retraining strategies Machine learning Cruz, Edmanuel Rangel, José Carlos Gomez Donoso, Francisco Cazorla, Miguel How to add new knowledge to already trained deep learning models applied to semantic localization |
description |
The capacity of a robot to automatically adapt to new environments is crucial, especially in social robotics. Often, when these robots are deployed in home or office environments, they tend to fail because they lack the ability to adapt to new and continuously changing scenarios. In order to accomplish this task, robots must obtain new information from the environment, and then add it to their already learned knowledge. Deep learning techniques are often used to tackle this problem successfully. However, these approaches, complete retraining of the models, which is highly time-consuming. In this work, several strategies are tested to find the best way to include new knowledge in an already learned model in a deep learning pipeline, putting the spotlight on the time spent for this training. We tackle the localization problem in the long term with a deep learning approach and testing several retraining strategies. The results of the experiments are discussed and, finally, the best approach is deployed on a Pepper robot. |
format |
Artículo |
author |
Cruz, Edmanuel Rangel, José Carlos Gomez Donoso, Francisco Cazorla, Miguel |
author_sort |
Cruz, Edmanuel |
title |
How to add new knowledge to already trained deep learning models applied to semantic localization |
title_short |
How to add new knowledge to already trained deep learning models applied to semantic localization |
title_full |
How to add new knowledge to already trained deep learning models applied to semantic localization |
title_fullStr |
How to add new knowledge to already trained deep learning models applied to semantic localization |
title_full_unstemmed |
How to add new knowledge to already trained deep learning models applied to semantic localization |
title_sort |
how to add new knowledge to already trained deep learning models applied to semantic localization |
publishDate |
2020 |
url |
https://link.springer.com/article/10.1007/s10489-019-01517-1 https://ridda2.utp.ac.pa/handle/123456789/9445 https://ridda2.utp.ac.pa/handle/123456789/9445 |
_version_ |
1796209477012160512 |
score |
12.2319145 |