Siamese capsule networks with global and local features for text classification

Wu, Yujia; Li, Jing; Wu, Jia; Chang, Jun

doi:10.1016/j.neucom.2020.01.064

Cited by 50 publications

(11 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…36,37 Among different one-shot learning techniques, Siamese Neural Network (SNN) 38 attracts much attention and provides state-of-art performance in image recognition 39 and bearing fault diagnosis. 40 Particularly, combining the CapsNets and SNNs, some researchers have proposed the Siamese CapsNets to solve the issues of face recognition 41 and text classification, 42 which can provide a good reference for this investigation. A problem that Siamese CapsNet faces is the extensive training parameters (high computation cost).…”

Section: The Proposed Methodsmentioning

confidence: 99%

A novel autonomous strategy for multi-bolt looseness detection using smart glove and Siamese double-path CapsNet

Wang

2021

Structural Health Monitoring

View full text Add to dashboard Cite

Recently, the issue of bolt looseness has attracted more attention due to its severe consequences. Among different methods for bolt looseness detection, the active sensing method that is based on stress wave signals is preferred since it is low cost and high robust. However, current active sensing method depends on permanent contact sensors, which may be impractical. Moreover, the investigation of multi-bolt looseness detection via the active sensing is very limited so far. With the above deficiency in mind, we propose a new robotic-assisted active sensing method based on our newly designed PZT-enabled smart gloves (SGs) and position-based visual servoing (PBVS) technique. Particularly, another main contribution is that we develop a new Siamese CapsNet to classify stress wave signals under different cases for multi-bolt looseness detection. Compared to machine learning (ML) and traditional deep learning techniques such as Convolutional Neural Networks (CNN), the proposed Siamese CapsNet model can achieve better performance and realize the recognition of signals that is never used during the training, which is impossible for common classification methods. Finally, an experiment is conducted to verify the effectiveness of the proposed method and Siamese CapsNet, which can guide future research significantly.

show abstract

Section: The Proposed Methodsmentioning

confidence: 99%

A novel autonomous strategy for multi-bolt looseness detection using smart glove and Siamese double-path CapsNet

Wang

2021

Structural Health Monitoring

View full text Add to dashboard Cite

show abstract

“…The work done in [26] utilized Siamese capsule networks as a tool for calculating the similarity of two short sequences of text via embeddings obtained from Bidirectional Gated Recurrent Units (BGRU). Employing a similar approach as the Siamese capsule network proposed for face recognition, the embeddings were first passed through primary capsules to identify parts of the text (a representation of words and phrases).…”

Section: Capsule Networkmentioning

confidence: 99%

Siamese Capsule Network for End-to-End Speaker Recognition in the Wild

Hajavi

Etemad

2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

We propose an end-to-end deep model for speaker verification in the wild. Our model uses thin-ResNet for extracting speaker embeddings from utterances and a Siamese capsule network and dynamic routing as the back-end to calculate a similarity score between the embeddings. We conduct a series of experiments and comparisons on our model versus baseline solutions, showing that our model outperforms the benchmarks using substantially less amount of training data. We also perform additional experiments to study the impact of different speaker embeddings on the Siamese capsule network. We show that the best performance is achieved by using embeddings obtained directly from the feature aggregation module of the front-end and passing them to higher capsules using dynamic routing.

show abstract

“…The convolutional layer of the capsule network is mainly used to extract local features. It uses capsule vectors to represent the spatial position relationship between local features [95]. To make up for the defect that the capsule network does not consider the context relationship of features, we have added an LSTM network before the capsule network to obtain context information of local features.…”

Section: Encoder Layermentioning

confidence: 99%

A Hybrid Siamese Neural Network for Natural Language Inference in Cyber-Physical Systems

et al. 2021

ACM Trans. Internet Technol.

View full text Add to dashboard Cite

Cyber-Physical Systems (CPS), as a multi-dimensional complex system that connects the physical world and the cyber world, has a strong demand for processing large amounts of heterogeneous data. These tasks also include Natural Language Inference (NLI) tasks based on text from different sources. However, the current research on natural language processing in CPS does not involve exploration in this field. Therefore, this study proposes a Siamese Network structure that combines Stacked Residual Long Short-Term Memory (bidirectional) with the Attention mechanism and Capsule Network for the NLI module in CPS, which is used to infer the relationship between text/language data from different sources. This model is mainly used to implement NLI tasks and conduct a detailed evaluation in three main NLI benchmarks as the basic semantic understanding module in CPS. Comparative experiments prove that the proposed method achieves competitive performance, has a certain generalization ability, and can balance the performance and the number of trained parameters.

show abstract

Siamese capsule networks with global and local features for text classification

Cited by 50 publications

References 8 publications

A novel autonomous strategy for multi-bolt looseness detection using smart glove and Siamese double-path CapsNet

A novel autonomous strategy for multi-bolt looseness detection using smart glove and Siamese double-path CapsNet

Siamese Capsule Network for End-to-End Speaker Recognition in the Wild

A Hybrid Siamese Neural Network for Natural Language Inference in Cyber-Physical Systems

Contact Info

Product

Resources

About