公开(公告)号：US09552549B1

公开(公告)日：2017-01-24

申请号：US14444272

申请日：2014-07-28

Applicant: Google Inc.

Inventor： Yunchao Gong ,  King Hong Thomas Leung ,  Alexander Toshkov Toshev ,  Sergey Ioffe ,  Yangqing Jia

IPC: G06N3/08

CPC classification number: G06N3/084 ,  G06N3/0454

Abstract: Systems and techniques are provided for a ranking approach to train deep neural nets for multilabel image annotation. Label scores may be received for labels determined by a neural network for training examples. Each label may be a positive label or a negative label for the training example. An error of the neural network may be determined based on a comparison, for each of the training examples, of the label scores for positive labels and negative labels for the training example and a semantic distance between each positive label and each negative label for the training example. Updated weights may be determined for the neural network based on a gradient of the determined error of the neural network. The updated weights may be applied to the neural network to train the neural network.

Abstract translation: 提供系统和技术用于排列方法来训练用于多标签图像注释的深层神经网络。 可以接收由用于训练示例的神经网络确定的标签的标签分数。 每个标签可能是培训示例的正标签或负标签。 可以基于针对训练样本的正标签的标签分数和训练样本的负标签的每个训练样本的比较以及训练样本的每个正标签和每个负标签之间的语义距离来确定神经网络的误差 例。 可以基于确定的神经网络的误差的梯度来确定神经网络的更新权重。 更新的权重可以应用于神经网络来训练神经网络。