• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.1
• /
• pp.1-14
• /
• 2024

Time-series forecasting is extensively used in the actual world. Recent research has shown that Transformers with a self-attention mechanism at their core exhibit better performance when dealing with such problems. However, most of the existing Transformer models used for time series prediction use the traditional encoder-decoder architecture, which is complex and leads to low model processing efficiency, thus limiting the ability to mine deep time dependencies by increasing model depth. Secondly, the secondary computational complexity of the self-attention mechanism also increases computational overhead and reduces processing efficiency. To address these issues, the paper designs an efficient multi-layer attention-based time-series forecasting model. This model has the following characteristics: (i) It abandons the traditional encoder-decoder based Transformer architecture and constructs a time series prediction model based on multi-layer attention mechanism, improving the model's ability to mine deep time dependencies. (ii) A cross attention module based on cross attention mechanism was designed to enhance information exchange between historical and predictive sequences. (iii) Applying a recently proposed sparse attention mechanism to our model reduces computational overhead and improves processing efficiency. Experiments on multiple datasets have shown that our model can significantly increase the performance of current advanced Transformer methods in time series forecasting, including LogTrans, Reformer, and Informer.

• ETRI Journal
• /
• v.42 no.1
• /
• pp.101-107
• /
• 2020

For Korean language processing, morphological analysis is a critical component that requires extensive work. This morphological analysis can be conducted in an end-to-end manner without requiring a complicated feature design using a sequence-to-sequence model. However, the sequence-to-sequence model has a time complexity of O(n²) for an input length n when using the attention mechanism technique for high performance. In this study, we propose a linear-time Korean morphological analysis model using a local monotonic attention mechanism relying on monotonic alignment, which is a characteristic of Korean morphological analysis. The proposed model indicates an extreme improvement in a single threaded environment and a high morphometric F1-measure even for a hard attention model with the elimination of the attention mechanism formula.

• ETRI Journal
• /
• v.43 no.5
• /
• pp.775-786
• /
• 2021

To translate in real time, a simultaneous translation system should determine when to stop reading source tokens and generate target tokens corresponding to a partial source sentence read up to that point. However, conventional attention-based neural machine translation (NMT) models cannot produce translations with adequate latency in online scenarios because they wait until a source sentence is completed to compute alignment between the source and target tokens. To address this issue, we propose a reinforced learning (RL)-based attention mechanism, the reinforced attention mechanism, which allows a neural translation model to jointly train the stopping criterion and a partial translation model. The proposed attention mechanism comprises two modules, one to ensure translation quality and the other to address latency. Different from previous RL-based simultaneous translation systems, which learn the stopping criterion from a fixed NMT model, the modules can be trained jointly with a novel reward function. In our experiments, the proposed model has better translation quality and comparable latency compared to previous models.

• Journal of the Institute of Convergence Signal Processing
• /
• v.22 no.4
• /
• pp.156-163
• /
• 2021

Attention mechanisms have been used in deep learning-based computer vision systems, including single image super-resolution (SISR) networks. However, existing SISR networks with attention mechanism focused on real image super-resolution, so it is hard to know whether they are available for CG or VR images. In this paper, we attempt to apply a recent attention module, called balanced attention mechanism (BAM) module, to 12 state-of-the-art SISR networks, and then check whether the BAM module can achieve performance improvement in CG or VR image super-resolution. In our experiments, it has been confirmed that the performance improvement in CG or VR image super-resolution is limited and depends on data characteristics, size, and network type.

• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.1
• /
• pp.8-14
• /
• 2022

With the development of deep learning, super-resolution (SR) methods have tried to use deep learning mechanism, instead of using simple interpolation. SR methods using deep learning is generally based on convolutional neural networks (CNN), but recently, SR researches using attention mechanism have been actively conducted. In this paper, we propose an approach of improving SR performance using one of the attention mechanisms, non-local sparse attention (NLSA). Through experiments, we confirmed that the performance of the existing SR models, IMDN, CARN, and OISR-LF-s can be improved by using NLSA.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.8
• /
• pp.2552-2570
• /
• 2022

Underwater images usually have various problems, such as the color cast of underwater images due to the attenuation of different lights in water, the darkness of image caused by the lack of light underwater, and the haze effect of underwater images because of the scattering of light. To address the above problems, the channel attention mechanism, strengthen-operate-subtract (SOS) boosting mechanism and gated fusion module are introduced in our paper, based on which, an underwater image recovery network is proposed. First, for the color cast problem of underwater images, the channel attention mechanism is incorporated in our model, which can well alleviate the color cast of underwater images. Second, as for the darkness of underwater images, the similarity between the target underwater image after dehazing and color correcting, and the image output by our model is used as the loss function, so as to increase the brightness of the underwater image. Finally, we employ the SOS boosting module to eliminate the haze effect of underwater images. Moreover, experiments were carried out to evaluate the performance of our model. The qualitative analysis results show that our method can be applied to effectively recover the underwater images, which outperformed most methods for comparison according to various criteria in the quantitative analysis.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.2
• /
• pp.544-564
• /
• 2022

Underwater images often suffer from color distortion, blurring and low contrast, which is caused by the propagation of light in the underwater environment being affected by the two processes: absorption and scattering. To cope with the poor quality of underwater images, this paper proposes a multiscale fusion underwater image enhancement method based on channel attention mechanism and local binary pattern (LBP). The network consists of three modules: feature aggregation, image reconstruction and LBP enhancement. The feature aggregation module aggregates feature information at different scales of the image, and the image reconstruction module restores the output features to high-quality underwater images. The network also introduces channel attention mechanism to make the network pay more attention to the channels containing important information. The detail information is protected by real-time superposition with feature information. Experimental results demonstrate that the method in this paper produces results with correct colors and complete details, and outperforms existing methods in quantitative metrics.

• Computers and Concrete
• /
• v.26 no.5
• /
• pp.411-420
• /
• 2020

Deep Convolution neural network (DCNN) has been widely used in the healthy maintenance of civil infrastructure. Using DCNN to improve crack detection performance has attracted many researchers' attention. In this paper, a light-weight spatial attention network module is proposed to strengthen the representation capability of ResNet and improve the crack detection performance. It utilizes attention mechanism to strengthen the interested objects in global receptive field of ResNet convolution layers. Global average spatial information over all channels are used to construct an attention scalar. The scalar is combined with adaptive weighted sigmoid function to activate the output of each channel's feature maps. Salient objects in feature maps are refined by the attention scalar. The proposed spatial attention module is stacked in ResNet50 to detect crack. Experiments results show that the proposed module can got significant performance improvement in crack detection.

• The Journal of the Acoustical Society of Korea
• /
• v.36 no.6
• /
• pp.407-412
• /
• 2017

In this paper, we propose a speech emotion recognition method using a deep neural network based on the attention mechanism. The proposed method consists of a combination of CNN (Convolution Neural Networks), GRU (Gated Recurrent Unit), DNN (Deep Neural Networks) and attention mechanism. The spectrogram of the speech signal contains characteristic patterns according to the emotion. Therefore, we modeled characteristic patterns according to the emotion by applying the tuned Gabor filters as convolutional filter of typical CNN. In addition, we applied the attention mechanism with CNN and FC (Fully-Connected) layer to obtain the attention weight by considering context information of extracted features and used it for emotion recognition. To verify the proposed method, we conducted emotion recognition experiments on six emotions. The experimental results show that the proposed method achieves higher performance in speech emotion recognition than the conventional methods.

• Journal of Information Processing Systems
• /
• v.17 no.2
• /
• pp.242-252
• /
• 2021

Following the intuition that the local information in time instances is hardly incorporated into the posterior sequence in long short-term memory (LSTM), this paper proposes an attention augmented mechanism for fault diagnosis of the complex chemical process data. Unlike conventional fault diagnosis and classification methods, an attention mechanism layer architecture is introduced to detect and focus on local temporal information. The augmented deep network results preserve each local instance's importance and contribution and allow the interpretable feature representation and classification simultaneously. The comprehensive comparative analyses demonstrate that the developed model has a high-quality fault classification rate of 95.49%, on average. The results are comparable to those obtained using various other techniques for the Tennessee Eastman benchmark process.