Visual recognition incorporating features of self-supervised models for the use of unlabelled data
Cargando...
Archivos
Fecha
2021
Autores
Profesor/a Guía
Facultad/escuela
Idioma
es
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Andrés Bello
Nombre de Curso
Licencia CC
Licencia CC
Resumen
Automatic visual object recognition has gained great popularity in the world and is
successfully applied to various areas such as robotics, security or commerce using deep
learning techniques. Training in machine learning models based on deep learning requires
an enormous amount of supervised data, which is expensive to obtain. An alternative is
to use semi-supervised models as co-training where the views given by deep networks are
differentiated using models that incorporate lateral information from each training object. In
this document, we describe and test a co-training model for deep networks, adding as auxiliary
inputs to self-supervised network features. The results show that the proposed model managed
to converge using a few dozen iterations, exceeding 2 % in precision compared to recent
models. This model, despite its simplicity, manages to be competitive with more complex
recent works. As future work, we plan to modify deep self-supervised networks to increase
diversity in co-training learning.
Notas
Tesis (Magíster en Ciencias de la Computación)
Palabras clave
Redes Neurales (Ciencia de la Computación), Algoritmos Computacionales