Method and device for data identification based on multitask deep neural network

Abstract

The invention discloses a method and a device for data identification based on a multitask deep neural network. The method comprises the following steps of: step 1, establishing the multitask deep neural network; step 2, regarding the two adjacent layers of the multitask deep neural network as a limited Boltzman machine, and pre-training the multitask deep neural network by a method of training layer by layer without supervision from bottom to top, so as to obtain an initial connection weight between the adjacent layers; step 3, minimizing a target function about the network weight with supervision by virtue of a back propagation algorithm, so as to obtain an optimized network weight; and step 4, inputting to-be-identified data in the multitask deep neural network with the optimized network weight, so as to obtain an output layer node value, thus obtaining the type of the to-be-identified data according to the output layer node value. In the method, relevance among different labels is excavated by virtue of the neural network, so that a high image labelling accuracy can still be ensured in large-scale image labelling with a high label quantity.

Description

technical field [0001] The invention relates to the field of pattern recognition and machine learning, in particular to a data recognition method and device based on a multi-task deep neural network. Background technique [0002] Multi-label learning problems widely exist in various aspects of real life. For example, in text classification, a text may contain multiple topics: health, medical and genetics. In natural scene classification, each scene may contain multiple categories: sky, beach, and ocean. Multi-label learning requires us to assign corresponding multiple labels to each text or image. [0003] Traditional binary and multi-class classification problems are special cases of multi-label learning, namely single-label learning problems. However, compared to assigning a single label to an instance, assigning multiple labels at the same time will inevitably increase the difficulty of solving the problem. The traditional solution is to solve the multi-label learning...

AI Technical Summary

Problems solved by technology

However, this method of learning each label independently does not consider the correlation between different labels, thus reducing the expressive ability of the model

Images