Baseline Needs More Love: On Simple Word-Embedding-Based Models and Associated Pooling Mechanisms
Dinghan Shen,Guoyin Wang,Wenlin Wang,Martin Renqiang Min,Qinliang Su,Yizhe Zhang,Chunyuan Li,Ricardo Henao,Lawrence Carin +8 more
- Vol. 1, pp 440-450
TLDR
This paper conducted 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.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.read more
Citations
More filters
Posted Content
Graph Convolutional Network with Generalized Factorized Bilinear Aggregation.
Hao Zhu,Piotr Koniusz +1 more
TL;DR: This article proposed a novel generalization of Factorized Bilinear (FB) layer to model the feature interactions in graph convolutional networks (GCNs) and proposed a compact FB layer by defining a family of summarizing operators applied over the quadratic term.
Journal ArticleDOI
Multiple userids identification with deep learning
TL;DR: Wang et al. as discussed by the authors used the deep learning model to extract deep semantic features at the document level and user level, then the most similar user pairs can be selected by computing the similarity of their writing styles indicated by text features.
Journal ArticleDOI
Heterogeneous Graph Neural Network for Short Text Classification
TL;DR: The TWPGCN model only focuses on the structural information of text graph, and does not require pre-training word embedding as initial node features, which improves classification accuracy, increases computational efficiency, and reduces computational difficulty.
Proceedings ArticleDOI
Deep Learning Models for Emotion Classification in Human Robot Interaction Platforms
TL;DR: In this paper , the authors compared the performance of common deep learning techniques (CNN, RNN) that could be used as a emotion classifier for HRI platforms such as chatbots or humanoid robots.
Proceedings ArticleDOI
Conceptual Grounding Constraints for Truly Robust Biomedical Name Representations
TL;DR: This article used conceptual grounding constraints to align encoded names to pre-trained embeddings of their concept identifiers, which is effective even when using a simple feed-forward encoding architecture that allows for scaling to large corpora while remaining sufficiently expressive.
References
More filters
Proceedings Article
Adam: A Method for Stochastic Optimization
Diederik P. Kingma,Jimmy Ba +1 more
TL;DR: This work introduces Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments, and provides a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework.
Journal ArticleDOI
Long short-term memory
TL;DR: A novel, efficient, gradient based method called long short-term memory (LSTM) is introduced, which can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units.
Proceedings Article
Attention is All you Need
Ashish Vaswani,Noam Shazeer,Niki Parmar,Jakob Uszkoreit,Llion Jones,Aidan N. Gomez,Lukasz Kaiser,Illia Polosukhin +7 more
TL;DR: This paper proposed a simple network architecture based solely on an attention mechanism, dispensing with recurrence and convolutions entirely and achieved state-of-the-art performance on English-to-French translation.
Journal Article
Dropout: a simple way to prevent neural networks from overfitting
TL;DR: It is shown that dropout improves the performance of neural networks on supervised learning tasks in vision, speech recognition, document classification and computational biology, obtaining state-of-the-art results on many benchmark data sets.
Proceedings ArticleDOI
Glove: Global Vectors for Word Representation
TL;DR: A new global logbilinear regression model that combines the advantages of the two major model families in the literature: global matrix factorization and local context window methods and produces a vector space with meaningful substructure.