Výsledky vyhledávání - "Recurrent neural network language models"

Report

Low-bit Quantization of Recurrent Neural Network Language Models Using Alternating Direction Methods of Multipliers

Autor: Xu, Junhao, Chen, Xie, Hu, Shoukang, Yu, Jianwei, Liu, Xunying, Meng, Helen

The high memory consumption and computational costs of Recurrent neural network language models (RNNLMs) limit their wider application on resource constrained devices. In recent years, neural network quantization techniques that are capable of produc

Externí odkaz: http://arxiv.org/abs/2111.14836

Zobrazit plný text záznamu

Report

Recurrent Neural Network Language Models Always Learn English-Like Relative Clause Attachment

Autor: Davis, Forrest, van Schijndel, Marten

A standard approach to evaluating language models analyzes how models assign probabilities to valid versus invalid syntactic constructions (i.e. is a grammatical sentence more probable than an ungrammatical sentence). Our work uses ambiguous relative

Externí odkaz: http://arxiv.org/abs/2005.00165

Zobrazit plný text záznamu

Report

Variational Smoothing in Recurrent Neural Network Language Models

Autor: Kong, Lingpeng, Melis, Gabor, Ling, Wang, Yu, Lei, Yogatama, Dani

We present a new theoretical perspective of data noising in recurrent neural network language models (Xie et al., 2017). We show that each variant of data noising is an instance of Bayesian recurrent neural networks with a particular variational dist

Externí odkaz: http://arxiv.org/abs/1901.09296

Zobrazit plný text záznamu

Report

Unsupervised and Efficient Vocabulary Expansion for Recurrent Neural Network Language Models in ASR

Autor: Khassanov, Yerbolat, Chng, Eng Siong

In automatic speech recognition (ASR) systems, recurrent neural network language models (RNNLM) are used to rescore a word lattice or N-best hypotheses list. Due to the expensive training, the RNNLM's vocabulary set accommodates only small shortlist

Externí odkaz: http://arxiv.org/abs/1806.10306

Zobrazit plný text záznamu

Report

Recurrent Neural Network Language Models for Open Vocabulary Event-Level Cyber Anomaly Detection

Autor: Tuor, Aaron, Baerwolf, Ryan, Knowles, Nicolas, Hutchinson, Brian, Nichols, Nicole, Jasper, Rob

Automated analysis methods are crucial aids for monitoring and defending a network to protect the sensitive or confidential data it hosts. This work introduces a flexible, powerful, and unsupervised approach to detecting anomalous behavior in compute

Externí odkaz: http://arxiv.org/abs/1712.00557

Zobrazit plný text záznamu

Report

BlackOut: Speeding up Recurrent Neural Network Language Models With Very Large Vocabularies

Autor: Ji, Shihao, Vishwanathan, S. V. N., Satish, Nadathur, Anderson, Michael J., Dubey, Pradeep

We propose BlackOut, an approximation algorithm to efficiently train massive recurrent neural network language models (RNNLMs) with million word vocabularies. BlackOut is motivated by using a discriminative loss, and we describe a new sampling strate

Externí odkaz: http://arxiv.org/abs/1511.06909

Zobrazit plný text záznamu

Report

Scaling Recurrent Neural Network Language Models

Autor: Williams, Will, Prasad, Niranjani, Mrva, David, Ash, Tom, Robinson, Tony

This paper investigates the scaling properties of Recurrent Neural Network Language Models (RNNLMs). We discuss how to train very large RNNs on GPUs and address the questions of how RNNLMs scale with respect to model size, training-set size, computat

Externí odkaz: http://arxiv.org/abs/1502.00512

Zobrazit plný text záznamu