检索结果-内蒙古大学图书馆

47th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

作者： Yang, Gene-Ping Tang, Hao Univ Edinburgh Ctr Speech Technol Res Sch Informat Edinburgh Midlothian Scotland

ISBN: (纸本)9781665405409

Attention mechanism in sequence-to-sequence models is designed to model the alignments between acoustic features and output tokens in speech recognition. However, attention weights produced by models trained end to end do not always correspond well with actual alignments, and several studies have further argued that attention weights might not even correspond well with the relevance attribution of frames. Regardless, visual similarity between attention weights and alignments is widely used during training as an indicator of the models quality. In this paper, we treat the correspondence between attention weights and alignments as a learning problem by imposing a supervised attention loss. Experiments have shown significant improved performance, suggesting that learning the alignments well during training critically determines the performance of sequence-to-sequence models.

关键词： Supervised attention Attention mechanism Speech recognition sequence-to-sequence models

来源：评论

学校读者我要写书评

暂无评论

Runoff predictions in ungauged basins using sequence-to-sequence models

引用

JOURNAL OF HYDROLOGY 2021年第PartB期603卷 126975-126975页

作者： Yin, Hanlin Guo, Zilong Zhang, Xiuwei Chen, Jiaojiao Zhang, Yanning Northwestern Polytech Univ Sch Comp Sci Shaanxi Key Lab Speech & Image Informat Proc SAII Xian 710072 Peoples R China Northwestern Polytech Univ Sch Comp Sci Natl Engn Lab Integrated AeroSp Ground Ocean Big Xian 710072 Peoples R China

How to improve the performance of runoff predictions in ungauged basins (PUB) is challenging. Recently, the long short-term memory (LSTM) based models have excellent performance and receive many attentions. In this paper, to improve the performance for 1-day-ahead runoff PUB and provide good performance for multi-day-ahead runoff PUB, we propose four sequence-to-sequence (S2S) models to deal with PUB. Furthermore, we introduce two modules named attribute-weighting module and multi-head-attention module for improving the performance. To show the power of these S2S models and the advantages of those two modules, we test our four S2S models and also three benchmark models including a PUB-LSTM model and two process-driven models by using k-fold validation on 531 basins of the Catchment Attributes and Meteorology for Large-Sample Studies (CAMELS) dataset. For 1-day-ahead runoff PUB, the median and the mean of Nash-Sutcliffe efficiency for the 531 basins provided by our best model achieve 0.78 and 0.70, respectively, while those provided by the PUB-LSTM model (the best one among those three benchmark models) are 0.69 and 0.54, respectively. Besides, our S2S models have promising results for multi-day-ahead runoff PUB. For 7th-day-ahead runoff PUB, the median and the mean of Nash-Sutcliffe efficiency for the 531 basins provided by our best model are 0.68 and 0.57, respectively. Furthermore, the results also show that the two modules are beneficial for improving the performance.

关键词： Rainfall-runoff modeling LSTM sequence-to-sequence models Ungauged basins

来源：评论

学校读者我要写书评

暂无评论

Empirical Evaluation of sequence-to-sequence models for Word Discovery in Low-resource Settings 20

Empirical Evaluation of Sequence-to-Sequence Models for Word...

引用

Interspeech Conference

作者： Boito, Marcely Zanon Villavicencio, Aline Besacier, Laurent Univ Grenoble Alpes UGA Lab Informat Grenoble Grenoble France Univ Essex Sch Comp Sci & Elect Engn Colchester Essex England Univ Fed Rio Grande do Sul Inst Informat Porto Alegre RS Brazil

Since Bahdanau et al. [1] first introduced attention for neural machine translation, most sequence-to-sequence models made use of attention mechanisms [2, 3, 4]. While they produce soft-alignment matrices that could be interpreted as alignment between target and source languages, we lack metrics to quantify their quality, being unclear which approach produces the best alignments. This paper presents an empirical evaluation of 3 of the main sequence-to-sequence models for word discovery from unsegmented phoneme sequences: CNN, RNN and Transformer-based. This task consists in aligning word sequences in a source language with phoneme sequences in a target language, inferring from it word segmentation on the target side [5]. Evaluating word segmentation quality can be seen as an extrinsic evaluation of the soft-alignment matrices produced during training. Our experiments in a low-resource scenario on Mboshi and English languages (both aligned to French) show that RNNs surprisingly outperform CNNs and Transformer for this task. Our results are confirmed by an intrinsic evaluation of alignment quality through the use Average Normalized Entropy (ANE). Lastly, we improve our best word discovery model by using an alignment entropy confidence measure that accumulates ANE over all the occurrences of a given alignment pair in the collection.

关键词： sequence-to-sequence models soft-alignment matrices word discovery low-resource languages computational language documentation

来源：评论

学校读者我要写书评

暂无评论

A detailed evaluation of neural sequence-to-sequence models for in-domain and cross-domain text simplification 11

A detailed evaluation of neural sequence-to-sequence models ...

引用

11th International Conference on Language Resources and Evaluation, LREC 2018

作者： Štajner, Sanja Nisioi, Sergiu DWS Research Group University of Mannheim Germany Human Language Technologies Research Center University of Bucharest Romania

ISBN: (纸本)9791095546009

We present a detailed evaluation and analysis of neural sequence-to-sequence models for text simplification on two distinct datasets: Wikipedia and Newsela. We employ both human and automatic evaluation to investigate the capacity of neural models to generalize across corpora, and we highlight challenges that these models face when tested on a different genre. Furthermore, we establish a strong baseline on the Newsela dataset and show that a simple neural architecture can be efficiently used for in-domain and cross-domain text simplification. © LREC 2018 - 11th International Conference on Language Resources and Evaluation. All rights reserved.

关键词： Evaluation Neural text simplification sequence-to-sequence models

来源：评论

学校读者我要写书评

暂无评论

Answer-Agnostic Question Generation in Privacy Policy Domain using sequence-to-sequence and Transformer models

Answer-Agnostic Question Generation in Privacy Policy Domain...

引用

作者： Deepti Lamba William H.Hsu Department of Computer Science Kansas State University

This paper presents a transformer and sequence-tosequence mapping approach,augmented using relevant named entities,towards generating questions for text understanding in the domain of policies(such as privacy agreements).To date,most work in question generation has used general-purpose benchmarks such as the SQuAD corpus;however,the application of question generation to document understanding has largely omitted use cases in the privacy policy domain.A privacy policy is a legal document that clearly divulges the policies of an organization regarding the gathering and usage of customer *** and understanding privacy policies to gain a measure of the rationality of terms of service before accepting them is of critical importance to ***,users frequently ignore these policies or base their decisions on insufficient understanding of the "fine print" due to the complex language and excessive length of *** work focuses on building a question generation system to assist reading comprehension of privacy *** work can be used to improve the quality of questionanswering systems in this domain,which consequently can help users understand the privacy policies before agreeing to *** paper uses existing deep learning models like T5,which is a transformer model,and several sequence-to-sequence models to generate ***,existing named entity recognition(NER) tools do not work in this domain,we also created our own named entity labels to add auxiliary information to our *** the auxiliary information improves the results over the baseline models giving us a promising future direction to further add some constraints,as a means to add more background knowledge to our models to generate better qualitv questions.

关键词： sequence-to-sequence models Privacy policy Question generation Transformer named entities

来源：评论

学校读者我要写书评

暂无评论

Evaluating sequence-to-sequence models for simulating medical staff mobility on time 7

Evaluating sequence-to-sequence models for simulating medica...

引用

7th Annual International Young Scientists Conference on Computational Science (YSC)

作者： Khovrichev, Mikhail A. Balakhontceva, Marina A. Ionov, Mikhail, V ITMO Univ eSci Res Inst 49 Kronverksky Pr St Petersburg 197101 Russia Almazov Natl Med Res Ctr 2 Akkuratova St St Petersburg 197341 Russia

The process of improving medical care effectiveness requires approaches for optimizing hospitality staff scheduling. Speaking of nursing care, their scheduling is directly connected with intra-hospital dynamics. In this paper, we present generator model for nurses' dynamics event chains based on encoder-decoder neural network. The model takes timestamp sequences with arbitrary lengths as input and produces sequences of places visited by simulated nurse. Our model is trained on physical access control system (PACS) data obtained from Almazov National Medical Research Centre. The number of methods for improving model accuracy were used. It was shown that trained generator produces realistic sequences with low level of hallucinations, opening new perspective of applications sequence-to-sequence models for system simulation. (C) 2018 The Authors. Published by Elsevier B. V. This is an open access article under the CC BY-NC-ND license (http://***/licenses/by-nc-nd/3.0/) Peer-review under responsibility of the scientific committee ofthe 7th International Young Scientist Conference on Computational Science.

关键词： neural netwoks sequence-to-sequence models event simulation medical staff mobility

来源：评论

学校读者我要写书评

暂无评论

MINIMUM WORD ERROR RATE TRAINING FOR ATTENTION-BASED sequence-to-sequence models

MINIMUM WORD ERROR RATE TRAINING FOR ATTENTION-BASED SEQUENC...

引用

IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

作者： Prabhavalkar, Rohit Sainath, Tara N. Wu, Yonghui Nguyen, Patrick Chen, Zhifeng Chiu, Chung-Cheng Kannan, Anjuli Google Inc Mountain View CA 94043 USA

ISBN: (纸本)9781538646588

sequence-to-sequence models, such as attention-based models in automatic speech recognition (ASR), are typically trained to optimize the cross-entropy criterion which corresponds to improving the log-likelihood of the data. However, system performance is usually measured in terms of word error rate (WER), not log-likelihood. Traditional ASR systems benefit from discriminative sequence training which optimizes criteria such as the state-level minimum Bayes risk (sMBR) which are more closely related to WER. In the present work, we explore techniques to train attention-based models to directly minimize expected word error rate. We consider two loss functions which approximate the expected number of word errors: either by sampling from the model, or by using N-best lists of decoded hypotheses, which we find to be more effective than the sampling-based method. In experimental evaluations, we find that the proposed training procedure improves performance by up to 8.2% relative to the baseline system. This allows us to train grapheme-based, uni-directional attention-based models which match the performance of a traditional, state-of-the-art, discriminative sequence-trained system on a mobile voice-search task.

关键词： sequence-to-sequence models attention models minimum word error rate training minimum Bayes risk

来源：评论

学校读者我要写书评

暂无评论

Evaluating sequence-to-sequence models for simulating medical staff mobility on time

引用

Procedia Computer Science 2018年 136卷 425-432页

作者： Mikhail A. Khovrichev Marina A. Balakhontceva Mikhail V. Ionov eScience Research Institute ITMO University 49 Kronverksky pr. St. Petersburg 197101 Russia Almazov National Medical Research Centre 2 Akkuratova st. St. Petersburg 197341 Russia

The process of improving medical care effectiveness requires approaches for optimizing hospitality staff scheduling. Speaking of nursing care, their scheduling is directly connected with intra-hospital dynamics. In this paper, we present generator model for nurses’ dynamics event chains based on encoder-decoder neural network. The model takes timestamp sequences with arbitrary lengths as input and produces sequences of places visited by simulated nurse. Our model is trained on physical access control system (PACS) data obtained from Almazov National Medical Research Centre. The number of methods for improving model accuracy were used. It was shown that trained generator produces realistic sequences with low level of hallucinations, opening new perspective of applications sequence-to-sequence models for system simulation.

关键词： neural netwoks sequence-to-sequence models event simulation medical staff mobility

来源：评论

学校读者我要写书评

暂无评论

MINIMUM WORD ERROR RATE TRAINING FOR ATTENTION-BASED sequence-to-sequence models

MINIMUM WORD ERROR RATE TRAINING FOR ATTENTION-BASED SEQUENC...

引用

IEEE International Conference on Acoustics, Speech and Signal Processing

作者： Rohit Prabhavalkar Tara N. Sainath Yonghui Wu Patrick Nguyen Zhifeng Chen Chung-Cheng Chiu Anjuli Kannan Google Inc

ISBN: (纸本)9781538646595

关键词： sequence-to-sequence models attention models minimum word error rate training minimum Bayes risk Speech recognition Error analysis LOSS FUNCTION Word aquifer storage and recovery System performance TRAIN Training training procedure

来源：评论

学校读者我要写书评

暂无评论

A Comparison of sequence-to-sequence models for Speech Recognition 18

A Comparison of Sequence-to-Sequence Models for Speech Recog...

引用

18th Annual Conference of the International-Speech-Communication-Association (INTERSPEECH 2017)

作者： Prabhavalkar, Rohit Rao, Kanishka Sainath, Tara N. Li, Bo Johnson, Leif Jaitly, Navdeep Google Inc Mountain View CA 94043 USA NVIDIA Santa Clara CA USA

ISBN: (纸本)9781510848764

In this work, we conduct a detailed evaluation of various all neural, end-to-end trained, sequence-to-sequence models applied to the task of speech recognition. Notably. each of these systems directly predicts graphemes in the written domain, without using an external pronunciation lexicon, or a separate language model. We examine several sequence-to-sequence models including connectionist temporal classification (CTC), the recurrent neural network (RNN) transducer, an attention based model, and a model which augments the RNN transducer with an attention mechanism. We find that the sequence-to-sequence models are competitive with traditional state-of-the-art approaches on dictation test sets, although the baseline, which uses a separate pronunciation and language model, outperforms these models on voice-search test sets.

关键词： sequence-to-sequence models attention models end-to-end models RNN transducer

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：