We propose a Topic Compositional Neural Language Model (TCNLM), a novel method designed to simultaneously capture both the global semantic meaning and the local word ordering structure in a document. The TCNLM learns the global semantic coherence of a document via a neural topic model, and the probability of each learned latent topic is further used to build a Mixture-of-Experts (MoE) language model, where each expert (corresponding to one topic) is a recurrent neural network (RNN) that accounts for learning the local structure of a word sequence. In order to train the MoE model efficiently, a matrix factorization method is applied, by extending each weight matrix of the RNN to be an ensemble of topic-dependent weight matrices. The degree to which each member of the ensemble is used is tied to the document-dependent probability of the corresponding topics. Experimental results on several corpora show that the proposed approach outperforms both a pure RNN-based model and other topic-guided language models. Further, our model yields sensible topics, and also has the capacity to generate meaningful sentences conditioned on given topics.

5
下载
关闭预览

相关内容

神经语言模型(Neural Language Model,NLM)是一类用来克服维数灾难的语言模型,它使用词的分布式表示对自然语言序列建模。不同于基于类的n-gram模型,神经语言模型在能够识别两个相似的词,并且不丧失将每个词编码为彼此不同的能力。神经语言模型共享一个词(及其上下文)和其他类似词。

Pre-trained deep neural network language models such as ELMo, GPT, BERT and XLNet have recently achieved state-of-the-art performance on a variety of language understanding tasks. However, their size makes them impractical for a number of scenarios, especially on mobile and edge devices. In particular, the input word embedding matrix accounts for a significant proportion of the model's memory footprint, due to the large input vocabulary and embedding dimensions. Knowledge distillation techniques have had success at compressing large neural network models, but they are ineffective at yielding student models with vocabularies different from the original teacher models. We introduce a novel knowledge distillation technique for training a student model with a significantly smaller vocabulary as well as lower embedding and hidden state dimensions. Specifically, we employ a dual-training mechanism that trains the teacher and student models simultaneously to obtain optimal word embeddings for the student vocabulary. We combine this approach with learning shared projection matrices that transfer layer-wise knowledge from the teacher model to the student model. Our method is able to compress the BERT_BASE model by more than 60x, with only a minor drop in downstream task metrics, resulting in a language model with a footprint of under 7MB. Experimental results also demonstrate higher compression efficiency and accuracy when compared with other state-of-the-art compression techniques.

0
17
下载
预览

In this paper, we propose Latent Relation Language Models (LRLMs), a class of language models that parameterizes the joint distribution over the words in a document and the entities that occur therein via knowledge graph relations. This model has a number of attractive properties: it not only improves language modeling performance, but is also able to annotate the posterior probability of entity spans for a given text through relations. Experiments demonstrate empirical improvements over both a word-based baseline language model and a previous approach that incorporates knowledge graph information. Qualitative analysis further demonstrates the proposed model's ability to learn to predict appropriate relations in context.

0
19
下载
预览

This paper proposes a state-of-the-art recurrent neural network (RNN) language model that combines probability distributions computed not only from a final RNN layer but also from middle layers. Our proposed method raises the expressive power of a language model based on the matrix factorization interpretation of language modeling introduced by Yang et al. (2018). The proposed method improves the current state-of-the-art language model and achieves the best score on the Penn Treebank and WikiText-2, which are the standard benchmark datasets. Moreover, we indicate our proposed method contributes to two application tasks: machine translation and headline generation. Our code is publicly available at: https://github.com/nttcslab-nlp/doc_lm.

0
5
下载
预览

This work focuses on combining nonparametric topic models with Auto-Encoding Variational Bayes (AEVB). Specifically, we first propose iTM-VAE, where the topics are treated as trainable parameters and the document-specific topic proportions are obtained by a stick-breaking construction. The inference of iTM-VAE is modeled by neural networks such that it can be computed in a simple feed-forward manner. We also describe how to introduce a hyper-prior into iTM-VAE so as to model the uncertainty of the prior parameter. Actually, the hyper-prior technique is quite general and we show that it can be applied to other AEVB based models to alleviate the {\it collapse-to-prior} problem elegantly. Moreover, we also propose HiTM-VAE, where the document-specific topic distributions are generated in a hierarchical manner. HiTM-VAE is even more flexible and can generate topic distributions with better variability. Experimental results on 20News and Reuters RCV1-V2 datasets show that the proposed models outperform the state-of-the-art baselines significantly. The advantages of the hyper-prior technique and the hierarchical model construction are also confirmed by experiments.

0
3
下载
预览

Humans can understand and produce new utterances effortlessly, thanks to their compositional skills. Once a person learns the meaning of a new verb "dax," he or she can immediately understand the meaning of "dax twice" or "sing and dax." In this paper, we introduce the SCAN domain, consisting of a set of simple compositional navigation commands paired with the corresponding action sequences. We then test the zero-shot generalization capabilities of a variety of recurrent neural networks (RNNs) trained on SCAN with sequence-to-sequence methods. We find that RNNs can make successful zero-shot generalizations when the differences between training and test commands are small, so that they can apply "mix-and-match" strategies to solve the task. However, when generalization requires systematic compositional skills (as in the "dax" example above), RNNs fail spectacularly. We conclude with a proof-of-concept experiment in neural machine translation, suggesting that lack of systematicity might be partially responsible for neural networks' notorious training data thirst.

0
3
下载
预览

Despite the effectiveness of recurrent neural network language models, their maximum likelihood estimation suffers from two limitations. It treats all sentences that do not match the ground truth as equally poor, ignoring the structure of the output space. Second, it suffers from "exposure bias": during training tokens are predicted given ground-truth sequences, while at test time prediction is conditioned on generated output sequences. To overcome these limitations we build upon the recent reward augmented maximum likelihood approach \ie sequence-level smoothing that encourages the model to predict sentences close to the ground truth according to a given performance metric. We extend this approach to token-level loss smoothing, and propose improvements to the sequence-level smoothing approach. Our experiments on two different tasks, image captioning and machine translation, show that token-level and sequence-level loss smoothing are complementary, and significantly improve results.

0
7
下载
预览

Topic models are one of the most frequently used models in machine learning due to its high interpretability and modular structure. However extending the model to include supervisory signal, incorporate pre-trained word embedding vectors and add nonlinear output function to the model is not an easy task because one has to resort to highly intricate approximate inference procedure. In this paper, we show that topic models could be viewed as performing a neighborhood aggregation algorithm where the messages are passed through a network defined over words. Under the network view of topic models, nodes corresponds to words in a document and edges correspond to either a relationship describing co-occurring words in a document or a relationship describing same word in the corpus. The network view allows us to extend the model to include supervisory signals, incorporate pre-trained word embedding vectors and add nonlinear output function to the model in a simple manner. Moreover, we describe a simple way to train the model that is well suited in a semi-supervised setting where we only have supervisory signals for some portion of the corpus and the goal is to improve prediction performance in the held-out data. Through careful experiments we show that our approach outperforms state-of-the-art supervised Latent Dirichlet Allocation implementation in both held-out document classification tasks and topic coherence.

0
3
下载
预览

Recurrent models for sequences have been recently successful at many tasks, especially for language modeling and machine translation. Nevertheless, it remains challenging to extract good representations from these models. For instance, even though language has a clear hierarchical structure going from characters through words to sentences, it is not apparent in current language models. We propose to improve the representation in sequence models by augmenting current approaches with an autoencoder that is forced to compress the sequence through an intermediate discrete latent space. In order to propagate gradients though this discrete representation we introduce an improved semantic hashing technique. We show that this technique performs well on a newly proposed quantitative efficiency measure. We also analyze latent codes produced by the model showing how they correspond to words and phrases. Finally, we present an application of the autoencoder-augmented model to generating diverse translations.

0
6
下载
预览

We propose a novel recommendation method based on tree. With user behavior data, the tree based model can capture user interests from coarse to fine, by traversing nodes top down and make decisions whether to pick up each node to user. Compared to traditional model-based methods like matrix factorization (MF), our tree based model does not have to fetch and estimate each item in the entire set. Instead, candidates are drawn from subsets corresponding to user's high-level interests, which is defined by the tree structure. Meanwhile, finding candidates from the entire corpus brings more novelty than content-based approaches like item-based collaborative filtering.Moreover, in this paper, we show that the tree structure can also act to refine user interests distribution, to benefit both training and prediction. The experimental results in both open dataset and Taobao display advertising dataset indicate that the proposed method outperforms existing methods.

0
7
下载
预览

Word Sense Disambiguation is an open problem in Natural Language Processing which is particularly challenging and useful in the unsupervised setting where all the words in any given text need to be disambiguated without using any labeled data. Typically WSD systems use the sentence or a small window of words around the target word as the context for disambiguation because their computational complexity scales exponentially with the size of the context. In this paper, we leverage the formalism of topic model to design a WSD system that scales linearly with the number of words in the context. As a result, our system is able to utilize the whole document as the context for a word to be disambiguated. The proposed method is a variant of Latent Dirichlet Allocation in which the topic proportions for a document are replaced by synset proportions. We further utilize the information in the WordNet by assigning a non-uniform prior to synset distribution over words and a logistic-normal prior for document distribution over synsets. We evaluate the proposed method on Senseval-2, Senseval-3, SemEval-2007, SemEval-2013 and SemEval-2015 English All-Word WSD datasets and show that it outperforms the state-of-the-art unsupervised knowledge-based WSD system by a significant margin.

0
5
下载
预览
小贴士
相关论文
Extreme Language Model Compression with Optimal Subwords and Shared Projections
Sanqiang Zhao,Raghav Gupta,Yang Song,Denny Zhou
17+阅读 · 2019年9月25日
Hiroaki Hayashi,Zecong Hu,Chenyan Xiong,Graham Neubig
19+阅读 · 2019年8月21日
Sho Takase,Jun Suzuki,Masaaki Nagata
5+阅读 · 2018年8月31日
Xuefei Ning,Yin Zheng,Zhuxi Jiang,Yu Wang,Huazhong Yang,Junzhou Huang
3+阅读 · 2018年6月18日
Maha Elbayad,Laurent Besacier,Jakob Verbeek
7+阅读 · 2018年5月14日
Ryohei Hisano
3+阅读 · 2018年2月22日
Łukasz Kaiser,Samy Bengio
6+阅读 · 2018年1月29日
Han Zhu,Pengye Zhang,Guozheng Li,Jie He,Han Li,Kun Gai
7+阅读 · 2018年1月8日
Devendra Singh Chaplot,Ruslan Salakhutdinov
5+阅读 · 2018年1月5日
相关VIP内容
【推荐系统/计算广告/机器学习/CTR预估资料汇总】
专知会员服务
72+阅读 · 2019年10月21日
强化学习最新教程,17页pdf
专知会员服务
74+阅读 · 2019年10月11日
机器学习入门的经验与建议
专知会员服务
47+阅读 · 2019年10月10日
Top