Co-Training for Visual Object Recognition Based on Self-Supervised Models Using a Cross-Entropy Regularization
Automatic recognition of visual objects using a deep learning approach has been successfully applied to multiple areas. However, deep learning techniques require a large amount of labeled data, which is usually expensive to obtain. An alternative is to use semi-supervised models, such as co-training...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-04-01
|
| Series: | Entropy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1099-4300/23/4/423 |
| _version_ | 1797539211396513792 |
|---|---|
| author | Gabriel Díaz Billy Peralta Luis Caro Orietta Nicolis |
| author_facet | Gabriel Díaz Billy Peralta Luis Caro Orietta Nicolis |
| author_sort | Gabriel Díaz |
| collection | DOAJ |
| description | Automatic recognition of visual objects using a deep learning approach has been successfully applied to multiple areas. However, deep learning techniques require a large amount of labeled data, which is usually expensive to obtain. An alternative is to use semi-supervised models, such as co-training, where multiple complementary views are combined using a small amount of labeled data. A simple way to associate views to visual objects is through the application of a degree of rotation or a type of filter. In this work, we propose a co-training model for visual object recognition using deep neural networks by adding layers of self-supervised neural networks as intermediate inputs to the views, where the views are diversified through the cross-entropy regularization of their outputs. Since the model merges the concepts of co-training and self-supervised learning by considering the differentiation of outputs, we called it Differential Self-Supervised Co-Training (DSSCo-Training). This paper presents some experiments using the DSSCo-Training model to well-known image datasets such as MNIST, CIFAR-100, and SVHN. The results indicate that the proposed model is competitive with the state-of-art models and shows an average relative improvement of 5% in accuracy for several datasets, despite its greater simplicity with respect to more recent approaches. |
| first_indexed | 2024-03-10T12:42:01Z |
| format | Article |
| id | doaj.art-dadab84e996744d2b0187c7794595b94 |
| institution | Directory Open Access Journal |
| issn | 1099-4300 |
| language | English |
| last_indexed | 2024-03-10T12:42:01Z |
| publishDate | 2021-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Entropy |
| spelling | doaj.art-dadab84e996744d2b0187c7794595b942023-11-21T13:48:18ZengMDPI AGEntropy1099-43002021-04-0123442310.3390/e23040423Co-Training for Visual Object Recognition Based on Self-Supervised Models Using a Cross-Entropy RegularizationGabriel Díaz0Billy Peralta1Luis Caro2Orietta Nicolis3Departamento de Ciencias de Ingeniería, Facultad de Ingeniería, Universidad Andres Bello, Antonio Varas 880, 8370146 Santiago, ChileDepartamento de Ciencias de Ingeniería, Facultad de Ingeniería, Universidad Andres Bello, Antonio Varas 880, 8370146 Santiago, ChileDepartamento de Ingeniería Informática, Facultad de Ingeniería, Universidad Católica de Temuco, Rudecindo Ortega 2950, 4781312 Temuco, ChileDepartamento de Ciencias de Ingeniería, Facultad de Ingeniería, Universidad Andres Bello, Antonio Varas 880, 8370146 Santiago, ChileAutomatic recognition of visual objects using a deep learning approach has been successfully applied to multiple areas. However, deep learning techniques require a large amount of labeled data, which is usually expensive to obtain. An alternative is to use semi-supervised models, such as co-training, where multiple complementary views are combined using a small amount of labeled data. A simple way to associate views to visual objects is through the application of a degree of rotation or a type of filter. In this work, we propose a co-training model for visual object recognition using deep neural networks by adding layers of self-supervised neural networks as intermediate inputs to the views, where the views are diversified through the cross-entropy regularization of their outputs. Since the model merges the concepts of co-training and self-supervised learning by considering the differentiation of outputs, we called it Differential Self-Supervised Co-Training (DSSCo-Training). This paper presents some experiments using the DSSCo-Training model to well-known image datasets such as MNIST, CIFAR-100, and SVHN. The results indicate that the proposed model is competitive with the state-of-art models and shows an average relative improvement of 5% in accuracy for several datasets, despite its greater simplicity with respect to more recent approaches.https://www.mdpi.com/1099-4300/23/4/423co-trainingdeep learningsemi-supervised learningself-supervised learning |
| spellingShingle | Gabriel Díaz Billy Peralta Luis Caro Orietta Nicolis Co-Training for Visual Object Recognition Based on Self-Supervised Models Using a Cross-Entropy Regularization Entropy co-training deep learning semi-supervised learning self-supervised learning |
| title | Co-Training for Visual Object Recognition Based on Self-Supervised Models Using a Cross-Entropy Regularization |
| title_full | Co-Training for Visual Object Recognition Based on Self-Supervised Models Using a Cross-Entropy Regularization |
| title_fullStr | Co-Training for Visual Object Recognition Based on Self-Supervised Models Using a Cross-Entropy Regularization |
| title_full_unstemmed | Co-Training for Visual Object Recognition Based on Self-Supervised Models Using a Cross-Entropy Regularization |
| title_short | Co-Training for Visual Object Recognition Based on Self-Supervised Models Using a Cross-Entropy Regularization |
| title_sort | co training for visual object recognition based on self supervised models using a cross entropy regularization |
| topic | co-training deep learning semi-supervised learning self-supervised learning |
| url | https://www.mdpi.com/1099-4300/23/4/423 |
| work_keys_str_mv | AT gabrieldiaz cotrainingforvisualobjectrecognitionbasedonselfsupervisedmodelsusingacrossentropyregularization AT billyperalta cotrainingforvisualobjectrecognitionbasedonselfsupervisedmodelsusingacrossentropyregularization AT luiscaro cotrainingforvisualobjectrecognitionbasedonselfsupervisedmodelsusingacrossentropyregularization AT oriettanicolis cotrainingforvisualobjectrecognitionbasedonselfsupervisedmodelsusingacrossentropyregularization |