Showing papers on "Word embedding published in 2018"

Universal Sentence Encoder

Abstract: We present models for encoding sentences into embedding vectors that specifically target transfer learning to other NLP tasks. The models are efficient and result in accurate performance on diverse transfer tasks. Two variants of the encoding models allow for trade-offs between accuracy and compute resources. For both variants, we investigate and report the relationship between model complexity, resource consumption, the availability of transfer task training data, and task performance. Comparisons are made with baselines that use word level transfer learning via pretrained word embeddings as well as baselines do not use any transfer learning. We find that transfer learning using sentence embeddings tends to outperform word level transfer. With transfer learning via sentence embeddings, we observe surprisingly good performance with minimal amounts of supervised training data for a transfer task. We obtain encouraging results on Word Embedding Association Tests (WEAT) targeted at detecting model bias. Our pre-trained sentence encoding models are made freely available for download and on TF Hub.

Contextual String Embeddings for Sequence Labeling

Abstract: Recent advances in language modeling using recurrent neural networks have made it viable to model language as distributions over characters. By learning to predict the next character on the basis of previous characters, such models have been shown to automatically internalize linguistic concepts such as words, sentences, subclauses and even sentiment. In this paper, we propose to leverage the internal states of a trained character language model to produce a novel type of word embedding which we refer to as contextual string embeddings. Our proposed embeddings have the distinct properties that they (a) are trained without any explicit notion of words and thus fundamentally model words as sequences of characters, and (b) are contextualized by their surrounding text, meaning that the same word will have different embeddings depending on its contextual use. We conduct a comparative evaluation against previous embeddings and find that our embeddings are highly useful for downstream tasks: across four classic sequence labeling tasks we consistently outperform the previous state-of-the-art. In particular, we significantly outperform previous work on English and German named entity recognition (NER), allowing us to report new state-of-the-art F1-scores on the CoNLL03 shared task. We release all code and pre-trained language models in a simple-to-use framework to the research community, to enable reproduction of these experiments and application of our proposed embeddings to other tasks: https://github.com/zalandoresearch/flair

Universal Sentence Encoder for English

Abstract: We present easy-to-use TensorFlow Hub sentence embedding models having good task transfer performance. Model variants allow for trade-offs between accuracy and compute resources. We report the relationship between model complexity, resources, and transfer performance. Comparisons are made with baselines without transfer learning and to baselines that incorporate word-level transfer. Transfer learning using sentence-level embeddings is shown to outperform models without transfer learning and often those that use only word-level transfer. We show good transfer task performance with minimal training data and obtain encouraging results on word embedding association tests (WEAT) of model bias.

Word embeddings quantify 100 years of gender and ethnic stereotypes.

Abstract: Word embeddings are a powerful machine-learning framework that represents each English word by a vector. The geometric relationship between these vectors captures meaningful semantic relationships between the corresponding words. In this paper, we develop a framework to demonstrate how the temporal dynamics of the embedding helps to quantify changes in stereotypes and attitudes toward women and ethnic minorities in the 20th and 21st centuries in the United States. We integrate word embeddings trained on 100 y of text data with the US Census to show that changes in the embedding track closely with demographic and occupation shifts over time. The embedding captures societal shifts-e.g., the women's movement in the 1960s and Asian immigration into the United States-and also illuminates how specific adjectives and occupations became more closely associated with certain populations over time. Our framework for temporal analysis of word embedding opens up a fruitful intersection between machine learning and quantitative social science.

Zero-Shot Recognition via Semantic Embeddings and Knowledge Graphs

Abstract: We consider the problem of zero-shot recognition: learning a visual classifier for a category with zero training examples, just using the word embedding of the category and its relationship to other categories, which visual data are provided. The key to dealing with the unfamiliar or novel category is to transfer knowledge obtained from familiar classes to describe the unfamiliar class. In this paper, we build upon the recently introduced Graph Convolutional Network (GCN) and propose an approach that uses both semantic embeddings and the categorical relationships to predict the classifiers. Given a learned knowledge graph (KG), our approach takes as input semantic embeddings for each node (representing visual category). After a series of graph convolutions, we predict the visual classifier for each category. During training, the visual classifiers for a few categories are given to learn the GCN parameters. At test time, these filters are used to predict the visual classifiers of unseen categories. We show that our approach is robust to noise in the KG. More importantly, our approach provides significant improvement in performance compared to the current state-of-the-art results (from 2 ~ 3% on some metrics to whopping 20% on a few).

Dissecting Contextual Word Embeddings: Architecture and Representation

Abstract: Contextual word representations derived from pre-trained bidirectional language models (biLMs) have recently been shown to provide significant improvements to the state of the art for a wide range of NLP tasks. However, many questions remain as to how and why these models are so effective. In this paper, we present a detailed empirical study of how the choice of neural architecture (e.g. LSTM, CNN, or self attention) influences both end task accuracy and qualitative properties of the representations that are learned. We show there is a tradeoff between speed and accuracy, but all architectures learn high quality contextual representations that outperform word embeddings for four challenging NLP tasks. Additionally, all architectures learn representations that vary with network depth, from exclusively morphological based at the word embedding layer through local syntax based in the lower contextual layers to longer range semantics such coreference at the upper layers. Together, these results suggest that unsupervised biLMs, independent of architecture, are learning much more about the structure of language than previously appreciated.

On the Limitations of Unsupervised Bilingual Dictionary Induction

Abstract: Unsupervised machine translation - i.e., not assuming any cross-lingual supervision signal, whether a dictionary, translations, or comparable corpora - seems impossible, but nevertheless, Lample et al. (2017) recently proposed a fully unsupervised machine translation (MT) model. The model relies heavily on an adversarial, unsupervised cross-lingual word embedding technique for bilingual dictionary induction (Conneau et al., 2017), which we examine here. Our results identify the limitations of current unsupervised MT: unsupervised bilingual dictionary induction performs much worse on morphologically rich languages that are not dependent marking, when monolingual corpora from different domains or different embedding algorithms are used. We show that a simple trick, exploiting a weak supervision signal from identical words, enables more robust induction and establish a near-perfect correlation between unsupervised bilingual dictionary induction performance and a previously unexplored graph similarity metric.

Generalizing and Improving Bilingual Word Embedding Mappings with a Multi-Step Framework of Linear Transformations.

Abstract: Using a dictionary to map independently trained word embeddings to a shared space has shown to be an effective approach to learn bilingual word embeddings. In this work, we propose a multi-step framework of linear transformations that generalizes a substantial body of previous work. The core step of the framework is an orthogonal transformation, and existing methods can be explained in terms of the additional normalization, whitening, re-weighting, de-whitening and dimensionality reduction steps. This allows us to gain new insights into the behavior of existing methods, including the effectiveness of inverse regression, and design a novel variant that obtains the best published results in zero-shot bilingual lexicon extraction. The corresponding software is released as an open source project.

Learning Gender-Neutral Word Embeddings

Abstract: Word embedding models have become a fundamental component in a wide range of Natural Language Processing (NLP) applications However, embeddings trained on human-generated corpora have been demonstrated to inherit strong gender stereotypes that reflect social constructs To address this concern, in this paper, we propose a novel training procedure for learning gender-neutral word embeddings Our approach aims to preserve gender information in certain dimensions of word vectors while compelling other dimensions to be free of gender influence Based on the proposed method, we generate a Gender-Neutral variant of GloVe (GN-GloVe) Quantitative and qualitative experiments demonstrate that GN-GloVe successfully isolates gender information without sacrificing the functionality of the embedding model

Dissecting Contextual Word Embeddings: Architecture and Representation

Abstract: Contextual word representations derived from pre-trained bidirectional language models (biLMs) have recently been shown to provide significant improvements to the state of the art for a wide range of NLP tasks. However, many questions remain as to how and why these models are so effective. In this paper, we present a detailed empirical study of how the choice of neural architecture (e.g. LSTM, CNN, or self attention) influences both end task accuracy and qualitative properties of the representations that are learned. We show there is a tradeoff between speed and accuracy, but all architectures learn high quality contextual representations that outperform word embeddings for four challenging NLP tasks. Additionally, all architectures learn representations that vary with network depth, from exclusively morphological based at the word embedding layer through local syntax based in the lower contextual layers to longer range semantics such coreference at the upper layers. Together, these results suggest that unsupervised biLMs, independent of architecture, are learning much more about the structure of language than previously appreciated.

Baseline Needs More Love: On Simple Word-Embedding-Based Models and Associated Pooling Mechanisms

Abstract: Many deep learning architectures have been proposed to model the compositionality in text sequences, requiring substantial number of parameters and expensive computations. However, there has not been a rigorous evaluation regarding the added value of sophisticated compositional functions. In this paper, we conduct a point-by-point comparative study between Simple Word-Embedding-based Models (SWEMs), consisting of parameter-free pooling operations, relative to word-embedding-based RNN/CNN models. Surprisingly, SWEMs exhibit comparable or even superior performance in the majority of cases considered. Based upon this understanding, we propose two additional pooling strategies over learned word embeddings: (i) a max-pooling operation for improved interpretability; and (ii) a hierarchical pooling operation, which preserves spatial (n-gram) information within text sequences. We present experiments on 17 datasets encompassing three tasks: (i) (long) document classification; (ii) text sequence matching; and (iii) short text tasks, including classification and tagging.

Gromov-Wasserstein Alignment of Word Embedding Spaces

Abstract: Cross-lingual or cross-domain correspondences play key roles in tasks ranging from machine translation to transfer learning. Recently, purely unsupervised methods operating on monolingual embeddings have become effective alignment tools. Current state-of-the-art methods, however, involve multiple steps, including heuristic post-hoc refinement strategies. In this paper, we cast the correspondence problem directly as an optimal transport (OT) problem, building on the idea that word embeddings arise from metric recovery algorithms. Indeed, we exploit the Gromov-Wasserstein distance that measures how similarities between pairs of words relate across languages. We show that our OT objective can be estimated efficiently, requires little or no tuning, and results in performance comparable with the state-of-the-art in various unsupervised word translation tasks.

Diachronic word embeddings and semantic shifts: a survey

Abstract: Recent years have witnessed a surge of publications aimed at tracing temporal changes in lexical semantics using distributional methods, particularly prediction-based word embedding models. However, this vein of research lacks the cohesion, common terminology and shared practices of more established areas of natural language processing. In this paper, we survey the current state of academic research related to diachronic word embeddings and semantic shifts detection. We start with discussing the notion of semantic shifts, and then continue with an overview of the existing methods for tracing such time-related shifts with word embedding models. We propose several axes along which these methods can be compared, and outline the main challenges before this emerging subfield of NLP, as well as prospects and possible applications.

Sentiment Polarity Detection for Software Development

Abstract: The role of sentiment analysis is increasingly emerging to study software developers' emotions by mining crowd-generated content within social software engineering tools. However, off-the-shelf sentiment analysis tools have been trained on non-technical domains and general-purpose social media, thus resulting in misclassifications of technical jargon and problem reports. Here, we present Senti4SD, a classifier specifically trained to support sentiment analysis in developers' communication channels. Senti4SD is trained and validated using a gold standard of Stack Overflow questions, answers, and comments manually annotated for sentiment polarity. It exploits a suite of both lexicon- and keyword-based features, as well as semantic features based on word embedding. With respect to a mainstream off-the-shelf tool, which we use as a baseline, Senti4SD reduces the misclassifications of neutral and positive posts as emotionally negative. To encourage replications, we release a lab package including the classifier, the word embedding space, and the gold standard with annotation guidelines.

The Geometry of Culture: Analyzing Meaning through Word Embeddings.

Abstract: We demonstrate the utility of a new methodological tool, neural-network word embedding models, for large-scale text analysis, revealing how these models produce richer insights into cultural associations and categories than possible with prior methods. Word embeddings represent semantic relations between words as geometric relationships between vectors in a high-dimensional space, operationalizing a relational model of meaning consistent with contemporary theories of identity and culture. We show that dimensions induced by word differences (e.g. man - woman, rich - poor, black - white, liberal - conservative) in these vector spaces closely correspond to dimensions of cultural meaning, and the projection of words onto these dimensions reflects widely shared cultural connotations when compared to surveyed responses and labeled historical data. We pilot a method for testing the stability of these associations, then demonstrate applications of word embeddings for macro-cultural investigation with a longitudinal analysis of the coevolution of gender and class associations in the United States over the 20th century and a comparative analysis of historic distinctions between markers of gender and class in the U.S. and Britain. We argue that the success of these high-dimensional models motivates a move towards "high-dimensional theorizing" of meanings, identities and cultural processes.

Learning Gender-Neutral Word Embeddings

Abstract: Word embedding models have become a fundamental component in a wide range of Natural Language Processing (NLP) applications. However, embeddings trained on human-generated corpora have been demonstrated to inherit strong gender stereotypes that reflect social constructs. To address this concern, in this paper, we propose a novel training procedure for learning gender-neutral word embeddings. Our approach aims to preserve gender information in certain dimensions of word vectors while compelling other dimensions to be free of gender influence. Based on the proposed method, we generate a Gender-Neutral variant of GloVe (GN-GloVe). Quantitative and qualitative experiments demonstrate that GN-GloVe successfully isolates gender information without sacrificing the functionality of the embedding model.

Diachronic word embeddings and semantic shifts: a survey

Abstract: Recent years have witnessed a surge of publications aimed at tracing temporal changes in lexical semantics using distributional methods, particularly prediction-based word embedding models. However, this vein of research lacks the cohesion, common terminology and shared practices of more established areas of natural language processing. In this paper, we survey the current state of academic research related to diachronic word embeddings and semantic shifts detection. We start with discussing the notion of semantic shifts, and then continue with an overview of the existing methods for tracing such time-related shifts with word embedding models. We propose several axes along which these methods can be compared, and outline the main challenges before this emerging subfield of NLP, as well as prospects and possible applications.

Gromov-Wasserstein Alignment of Word Embedding Spaces.

Abstract: Cross-lingual or cross-domain correspondences play key roles in tasks ranging from machine translation to transfer learning. Recently, purely unsupervised methods operating on monolingual embeddings have become effective alignment tools. Current state-of-the-art methods, however, involve multiple steps, including heuristic post-hoc refinement strategies. In this paper, we cast the correspondence problem directly as an optimal transport (OT) problem, building on the idea that word embeddings arise from metric recovery algorithms. Indeed, we exploit the Gromov-Wasserstein distance that measures how similarities between pairs of words relate across languages. We show that our OT objective can be estimated efficiently, requires little or no tuning, and results in performance comparable with the state-of-the-art in various unsupervised word translation tasks.

On the dimensionality of word embedding

Abstract: In this paper, we provide a theoretical understanding of word embedding and its dimensionality. Motivated by the unitary-invariance of word embedding, we propose the Pairwise Inner Product (PIP) loss, a novel metric on the dissimilarity between word embeddings. Using techniques from matrix perturbation theory, we reveal a fundamental bias-variance trade-off in dimensionality selection for word embeddings. This bias-variance trade-off sheds light on many empirical observations which were previously unexplained, for example the existence of an optimal dimensionality. Moreover, new insights and discoveries, like when and how word embeddings are robust to over-fitting, are revealed. By optimizing over the bias-variance trade-off of the PIP loss, we can explicitly answer the open question of dimensionality selection for word embedding.

Botnet Detection in the Internet of Things using Deep Learning Approaches

Abstract: The recent growth of the Internet of Things (IoT) has resulted in a rise in IoT based DDoS attacks. This paper presents a solution to the detection of botnet activity within consumer IoT devices and networks. A novel application of Deep Learning is used to develop a detection model based on a Bidirectional Long Short Term Memory based Recurrent Neural Network (BLSTM-RNN). Word Embedding is used for text recognition and conversion of attack packets into tokenised integer format. The developed BLSTM-RNN detection model is compared to a LSTM-RNN for detecting four attack vectors used by the mirai botnet, and evaluated for accuracy and loss. The paper demonstrates that although the bidirectional approach adds overhead to each epoch and increases processing time, it proves to be a better progressive model over time. A labelled dataset was generated as part of this research, and is available upon request.

Semantic Sentence Matching with Densely-connected Recurrent and Co-attentive Information

Abstract: Sentence matching is widely used in various natural language tasks such as natural language inference, paraphrase identification, and question answering. For these tasks, understanding logical and semantic relationship between two sentences is required but it is yet challenging. Although attention mechanism is useful to capture the semantic relationship and to properly align the elements of two sentences, previous methods of attention mechanism simply use a summation operation which does not retain original features enough. Inspired by DenseNet, a densely connected convolutional network, we propose a densely-connected co-attentive recurrent neural network, each layer of which uses concatenated information of attentive features as well as hidden features of all the preceding recurrent layers. It enables preserving the original and the co-attentive feature information from the bottommost word embedding layer to the uppermost recurrent layer. To alleviate the problem of an ever-increasing size of feature vectors due to dense concatenation operations, we also propose to use an autoencoder after dense concatenation. We evaluate our proposed architecture on highly competitive benchmark datasets related to sentence matching. Experimental results show that our architecture, which retains recurrent and attentive features, achieves state-of-the-art performances for most of the tasks.

Neural information retrieval: at the end of the early years

Abstract: A recent “third wave” of neural network (NN) approaches now delivers state-of-the-art performance in many machine learning tasks, spanning speech recognition, computer vision, and natural language processing. Because these modern NNs often comprise multiple interconnected layers, work in this area is often referred to as deep learning. Recent years have witnessed an explosive growth of research into NN-based approaches to information retrieval (IR). A significant body of work has now been created. In this paper, we survey the current landscape of Neural IR research, paying special attention to the use of learned distributed representations of textual units. We highlight the successes of neural IR thus far, catalog obstacles to its wider adoption, and suggest potentially promising directions for future research.

Interpretable Adversarial Perturbation in Input Embedding Space for Text

Abstract: Following great success in the image processing field, the idea of adversarial training has been applied to tasks in the natural language processing (NLP) field. One promising approach directly applies adversarial training developed in the image processing field to the input word embedding space instead of the discrete input space of texts. However, this approach abandons such interpretability as generating adversarial texts to significantly improve the performance of NLP tasks. This paper restores interpretability to such methods by restricting the directions of perturbations toward the existing words in the input embedding space. As a result, we can straightforwardly reconstruct each input with perturbations to an actual text by considering the perturbations to be the replacement of words in the sentence while maintaining or even improving the task performance.

Style Transfer as Unsupervised Machine Translation

Abstract: Language style transferring rephrases text with specific stylistic attributes while preserving the original attribute-independent content. One main challenge in learning a style transfer system is a lack of parallel data where the source sentence is in one style and the target sentence in another style. With this constraint, in this paper, we adapt unsupervised machine translation methods for the task of automatic style transfer. We first take advantage of style-preference information and word embedding similarity to produce pseudo-parallel data with a statistical machine translation (SMT) framework. Then the iterative back-translation approach is employed to jointly train two neural machine translation (NMT) based transfer systems. To control the noise generated during joint training, a style classifier is introduced to guarantee the accuracy of style transfer and penalize bad candidates in the generated pseudo data. Experiments on benchmark datasets show that our proposed method outperforms previous state-of-the-art models in terms of both accuracy of style transfer and quality of input-output correspondence.

Unsupervised Cross-lingual Transfer of Word Embedding Spaces.

Abstract: Cross-lingual transfer of word embeddings aims to establish the semantic mappings among words in different languages by learning the transformation functions over the corresponding word embedding spaces Successfully solving this problem would benefit many downstream tasks such as to translate text classification models from resource-rich languages (eg English) to low-resource languages Supervised methods for this problem rely on the availability of cross-lingual supervision, either using parallel corpora or bilingual lexicons as the labeled data for training, which may not be available for many low resource languages This paper proposes an unsupervised learning approach that does not require any cross-lingual labeled data Given two monolingual word embedding spaces for any language pair, our algorithm optimizes the transformation functions in both directions simultaneously based on distributional matching as well as minimizing the back-translation losses We use a neural network implementation to calculate the Sinkhorn distance, a well-defined distributional similarity measure, and optimize our objective through back-propagation Our evaluation on benchmark datasets for bilingual lexicon induction and cross-lingual word similarity prediction shows stronger or competitive performance of the proposed method compared to other state-of-the-art supervised and unsupervised baseline methods over many language pairs

Deep Learning Approach for Short-Term Stock Trends Prediction Based on Two-Stream Gated Recurrent Unit Network

Abstract: Financial news has been proven to be a crucial factor which causes fluctuations in stock prices. However, previous studies heavily relied on analyzing shallow features and ignored the structural relation among words in a sentence. Several sentiment analysis studies have tried to point out the relationship between investors’ reaction and news events. However, the sentiment dataset was usually constructed from the lingual dataset which is unrelated to the financial sector and led to poor performance. This paper proposes a novel framework to predict the directions of stock prices by using both financial news and sentiment dictionary. The original contributions of this paper include the proposal of a novel two-stream gated recurrent unit network and Stock2Vec—a sentiment word embedding trained on financial news dataset and Harvard IV-4. Two main experiments are conducted: the first experiment predicts SP 2) Stock2Vec is more efficient in dealing with financial datasets; and 3) applying the model, a simulation scenario proves that our model is effective for the stock sector.

Refining Word Embeddings Using Intensity Scores for Sentiment Analysis

Abstract: Word embeddings that provide continuous low-dimensional vector representations of words have been extensively used for various natural language processing tasks. However, existing context-based word embeddings such as Word2vec and GloVe typically fail to capture sufficient sentiment information, which may result in words with similar vector representations having an opposite sentiment polarity (e.g., good and bad ), thus degrading sentiment analysis performance. To tackle this problem, recent studies have suggested learning sentiment embeddings to incorporate the sentiment polarity (positive and negative) information from labeled corpora. This study adopts another strategy to learn sentiment embeddings. Instead of creating a new word embedding from labeled corpora, we propose a word vector refinement model to refine existing pretrained word vectors using real-valued sentiment intensity scores provided by sentiment lexicons. The idea of the refinement model is to improve each word vector such that it can be closer in the lexicon to both semantically and sentimentally similar words (i.e., those with similar intensity scores) and further away from sentimentally dissimilar words (i.e., those with dissimilar intensity scores). An obvious advantage of the proposed method is that it can be applied to any pretrained word embeddings. In addition, the intensity scores can provide more fine-grained (real-valued) sentiment information than binary polarity labels to guide the refinement process. Experimental results show that the proposed refinement model can improve both conventional word embeddings and previously proposed sentiment embeddings for binary, ternary, and fine-grained sentiment classification on the SemEval and Stanford Sentiment Treebank datasets.

OntoSenticNet: A Commonsense Ontology for Sentiment Analysis

Abstract: In this work, we present OntoSenticNet, a commonsense ontology for sentiment analysis based on SenticNet, a semantic network of 100,000 concepts based on conceptual primitives. The key characteristics of OntoSenticNet are: (i) the definition of precise conceptual hierarchy and properties associating concepts and sentiment values; (ii) the support for connecting external information (e.g., word embedding, domain information, and different polarity representations) to each individual defined within the ontology; and (iii) the capability of associating each concept with annotations contained in external resources (e.g., documents and multimodal resources).

Retrieval on source code: a neural code search

Abstract: Searching over large code corpora can be a powerful productivity tool for both beginner and experienced developers because it helps them quickly find examples of code related to their intent. Code search becomes even more attractive if developers could express their intent in natural language, similar to the interaction that Stack Overflow supports. In this paper, we investigate the use of natural language processing and information retrieval techniques to carry out natural language search directly over source code, i.e. without having a curated Q&A forum such as Stack Overflow at hand. Our experiments using a benchmark suite derived from Stack Overflow and GitHub repositories show promising results. We find that while a basic word–embedding based search procedure works acceptably, better results can be obtained by adding a layer of supervision, as well as by a customized ranking strategy.

Using a stacked residual LSTM model for sentiment intensity prediction

Abstract: The sentiment intensity of a text indicates the strength of its association with positive sentiment, which is a continuous real-value between 0 and 1. Compared to polarity classification, predicting sentiment intensities for texts can achieve more fine-grained sentiment analyses. By introducing word embedding techniques, recent studies that use deep neural models have outperformed existing lexicon- and regression-based methods for sentiment intensity prediction. For better performance, a common way of a neural network is to add more layers in order to learn high-level features. However, when the depth increases, the network degrades and becomes more difficult to train. Since the errors between layers will be accumulated, and gradients will be vanished. To address this problem, this paper proposes a stacked residual LSTM model to predict sentiment intensity for a given text. By investigating the performances of shallow and deep architectures, we introduce a residual connection to every few LSTM layers to construct an 8-layer neural network. The residual connection can center layer gradients and propagated errors. Thus it makes the deeper network easier to optimize. This approach enables us to stack more layers of LSTM successfully for this task, which can improve the prediction accuracy of existing methods. Experimental results show that the proposed method outperforms lexicon-, regression-, and conventional NN-based methods proposed in previous studies.