Co-Training for Visual Object Recognition Based on Self-Supervised Models Using a Cross-Entropy Regularization
| dc.contributor.author | Diaz, Gabriel | |
| dc.contributor.author | Peralta, Billy | |
| dc.contributor.author | Caro, Luis | |
| dc.contributor.author | Nicolis, Orietta | |
| dc.date | 2021 | |
| dc.date.accessioned | 2021-10-04T18:54:32Z | |
| dc.date.available | 2021-10-04T18:54:32Z | |
| dc.description.abstract | 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. | |
| dc.identifier.citation | ENTROPY,Vol.23,,2021 | |
| dc.identifier.doi | 10.3390/e23040423 | |
| dc.identifier.uri | http://repositoriodigital.uct.cl/handle/10925/4349 | |
| dc.language.iso | en | |
| dc.publisher | MDPI | |
| dc.source | ENTROPY | |
| dc.subject.english | co-training | |
| dc.subject.english | deep learning | |
| dc.subject.english | semi-supervised learning | |
| dc.subject.english | self-supervised learning | |
| dc.subject.english | ||
| dc.title | Co-Training for Visual Object Recognition Based on Self-Supervised Models Using a Cross-Entropy Regularization | |
| dc.type | Article | |
| uct.indizacion | SCI |