公开(公告)号：US20140164299A1

公开(公告)日：2014-06-12

申请号：US13707088

申请日：2012-12-06

申请人： Tara Sainath ,  Brian Kingsbury ,  Bhuvana Ramabhadran

发明人： Tara Sainath ,  Brian Kingsbury ,  Bhuvana Ramabhadran

IPC分类号： G06N3/08

CPC分类号： G06N3/08

摘要： Pretraining for a DBN initializes weights of the DBN (Deep Belief Network) using a hybrid pre-training methodology. Hybrid pre-training employs generative component that allows the hybrid PT method to have better performance in WER (Word Error Rate) compared to the discriminative PT method. Hybrid pre-training learns weights which are more closely linked to the final objective function, allowing for a much larger batch size compared to generative PT, which allows for improvements in speed; and a larger batch size allows for parallelization of the gradient computation, speeding up training further.

摘要翻译： 预先训练DBN使用混合预训练方法初始化DBN（深信仰网络）的权重。 混合预训练采用生成部件，与辨别性PT方法相比，允许混合PT方法在WER（字错误率）方面具有更好的性能。 混合预训练学习与最终目标函数更紧密相关的权重，允许与生成PT相比更大的批量大小，这允许提高速度; 并且较大的批量允许梯度计算的并行化，进一步加速训练。