Semantic Compositionality through Recursive Matrix-Vector Spaces

作者信息:
Richard Socher Brody Huval Christopher D. Manning Andrew Y. Ng
richard@socher.org, {brodyh,manning,ang}@stanford.edu
Computer Science Department, Stanford University
代码数据公开:
https://www.socher.org/index.php/Main/

MX-RNN模型:
We present a novel recursive neural network model for semantic compositionality. In our context,compositionality is the ability to learn compositional vector representations for various types of phrases and sentences of arbitrary length.
不需要任何hand-designed semantic resources比如WorNet,FrameNet
通过加入WordNet hypernyms, POS and NER tags性能更好

2 MV-RNN: A Recursive Matrix-Vector Model
之前多个词向量组合成一个短语或者句子是用的线性关系,但是只能捕捉‘sum’的关系,不能捕捉可以改变另外一个词的含义的词的功能,例如extremely strong
1)给每个词分配了一个向量和一个矩阵
2) learning an input-specific, nonlinear, compositional function for computing vector and matrix representations for multi-word sequences of any syntactic type
如果一个词缺少operator semantics,那么它的矩阵就是一个单位矩阵;然而,如果一个词主要是作为operator,比如extremely,那么它的向量会接近0,它的矩阵gain a clear operator meaning,正向和负向都会增大被修饰词的含义

2.2 Matrix-Vector Neural Word Representation
与训练词向量为50维,矩阵50*50维,初始化矩阵是单位矩阵+高斯噪声,所以每个句子表示为((单词向量1,单词矩阵1),。。。,(),())

2.2 composition models for two words
2010年的工作,用了表示句法关系的矩阵R,背景知识的矩阵K
我们的工作:u需要任何手工涉及的semantic resources比如背景知识K,也不需要explicit knownledge of relation R,使用输入independant的组合函数
W是n*2n,可以将输入的词统一到相同的维度,W可以捕捉compositional信息
得到的是p

2.3 Recursive Compositions of Multiple Words and Phrases
This section describes how we extend a word-pair matrix-vector-based compositional model to learn vectors and matrices for longer sequences of words
Wm
得到的是P

2.4 Objective Functions for Training
One of the advantages of RNN-based models is that each node of a tree has associated with it a distributed vector representation (the parent vector p) which can also be seen as features describing that
phrase.
softmax对节点p分类

2.5learning

2.6 Low-Rank Matrix Approximations

5 combination od semantic relationship
semantic relationships between pairs of nominals. For instance, in the sentence “My [apartment]e1 has a pretty large [kitchen]e2.”, we want to predict that the kitchen and apartment are in a component-whole relationship.
figure5分类名词关系:
1)先找到要分类的两个词
2)使用节点向量分类
数据是9个顺序关系(两个方向)*2+没有方向的一个 = 19类,比如因果,文本-主题

Semantic Compositionality through Recursive Matrix-Vector Spaces-paper的更多相关文章

  1. 论文翻译——Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank

    Abstract Semantic word spaces have been very useful but cannot express the meaning of longer phrases ...

  2. 向量空间(Vector Spaces)

    向量空间(Vector Spaces) 向量空间又称线性空间,是线性代数的中心内容和基本概念之一.在解析几何里引入向量的概念后,是许多问题的处理变得更为简洁和清晰,在此基础上的进一步抽象化,形成了与域 ...

  3. 语义SLAM的数据关联和语义定位(二)Semantic Localization Via the Matrix Permanent

    论文假设和单目标模型 这部分想讲一下Semantic Localization Via the Matrix Permanent这篇文章的一些假设. 待求解的问题可以描述为 假设从姿态\(x\)看到的 ...

  4. 【读书笔记】:MIT线性代数(2):Vector Spaces and Subspaces

    Vector Space: R1, R2, R3,R4 , .... Each space Rn consists of a whole collection of vectors. R5 conta ...

  5. Deep Learning for NLP 文章列举

    Deep Learning for NLP 文章列举 原文链接:http://www.xperseverance.net/blogs/2013/07/2124/   大部分文章来自: http://w ...

  6. 转 Deep Learning for NLP 文章列举

    原文链接:http://www.xperseverance.net/blogs/2013/07/2124/   大部分文章来自: http://www.socher.org/ http://deepl ...

  7. tree-lstm初探

    https://zhuanlan.zhihu.com/p/35252733 可以先看看上面知乎文章里面的例子 Socher 等人于2012和2013年分别提出了两种区分词或短语类型的模型,即SU-RN ...

  8. 【MT】牛津的MT教程

    Preamble This repository contains the lecture slides and course description for the Deep Natural Lan ...

  9. Saw a tweet from Andrew Liam Trask, sounds like Oxford DeepNLP 2017 class have all videos slides practicals all up. Thanks Andrew for the tip!

    Saw a tweet from Andrew Liam Trask, sounds like Oxford DeepNLP 2017 class have all videos/slides/pra ...

随机推荐

  1. Linux PWM framework简介和API描述【转】

    本文转载自:https://blog.csdn.net/mike8825/article/details/51656400 1. 前言 PWM是Pulse Width Modulation(脉冲宽度调 ...

  2. 遗留系统如何用DDD重构(未完,待续)

    GETTING DDD STARTED SURROUNDED BY LEGACY SYSTEMS ByEric Evans

  3. Static需谨慎

    Static Cling Sticking Your Code To Things Unnecessarily Static Cling is a code smell used to describ ...

  4. ORM模型

    一.创建及映射(orm_intro_demo文件) 在项目新建App下的models.py文件下新建ORM模型: from django.db import models #如果要将一个普通的类变成一 ...

  5. 【NET Core】.NET Core中读取json配置文件

    在.NET Framework框架下应用配置内容一般都是写在Web.config或者App.config文件中,读取这两个配置文件只需要引用System.Configuration程序集,分别用 Sy ...

  6. DAY8 文件操作(二)

    一.写 1.1写文件 # w:没有文件新建文件,有文件就清空文件 w = open('1.txt', 'w', encoding='utf-8') w.write('000\n') # 在写入大量数据 ...

  7. JS封装addClass、removeClass

    addClass封装:1.先把原有的类名和需要添加的类名用“”切割.拼接. 2.查重,把所有类名遍历,重复的去掉. 3.“”拼接. function addClass(ele , cName) { v ...

  8. laravel 路由的配置

  9. 小程序 input 组件内容显示不全(显示的长度不满 input 宽度)问题

    问题:小程序的input组件经常用到,但在使用input组件的时候会出现一种现象:明明设置了input的宽度,但是输入的内容显示的长度范围却怎么都不到一整个input组件的宽度,而且后面没显示的地方无 ...

  10. scrapy python2升级python3遇到的坑

    换成Python3首先pycharm先执行: 然后看代码自己所需要的第三方库都要重新装 然后执行代码: 遇到这样的错如下: SyntaxError: invalid syntax 先检查print 所 ...