摘要 :
Noise cancelation and system identification have been studied for many years, an adaptive filters have proved to be a good means for solving such problems. Some neural net- works can be treated as nonlinear adaptive filters, and r...
展开
Noise cancelation and system identification have been studied for many years, an adaptive filters have proved to be a good means for solving such problems. Some neural net- works can be treated as nonlinear adaptive filters, and re thus expected to be more powerful than traditional adaptive filters when dealing with nonlinear system problems. In this paper, two new heterogeneous recurrent neural network (HRNN) archi- tectures will be proposed to identify some nonlinear systems and to extract a fetal electrocardiogram (ECG), which is corrupted by a much larger noise signal, Mother's ECG.
收起
摘要 :
This paper presents a design method for noise cancellation using recurrent neural network (RNN). When the observation signal is composed of the sum of the noise with the desired signal, it is a fundamental and important problem to...
展开
This paper presents a design method for noise cancellation using recurrent neural network (RNN). When the observation signal is composed of the sum of the noise with the desired signal, it is a fundamental and important problem to reduce the noise in speech recognition, etc.. There is the noise canceller as the tool which removes the noise. The conventional noise canceller needs the reference input terminal. In this paper, the technique by which noise reduction is possible using RNN and the observation signal only, is proposed. Finally, the results of computer simulation are presented to illustrate the effectiveness of the proposed method.
收起
摘要 :
This paper presents a design method for noise cancellation using recurrent neural network (RNN). When the observation signal is composed of the sum of the noise with the desired signal, it is a fundamental and important problem to...
展开
This paper presents a design method for noise cancellation using recurrent neural network (RNN). When the observation signal is composed of the sum of the noise with the desired signal, it is a fundamental and important problem to reduce the noise in speech recognition, etc.. There is the noise canceller as the tool which removes the noise. The conventional noise canceller needs the reference input terminal. In this paper, the technique by which noise reduction is possible using RNN and the observation signal only, is proposed. Finally, the results of computer simulation are presented to illustrate the effectiveness of the proposed method.
收起
摘要 :
This paper presents a design method for noise cancellation using recurrent neural network (RNN). When the observation signal is composed of the sum of the noise with the desired signal, it is a fundamental and important problem to...
展开
This paper presents a design method for noise cancellation using recurrent neural network (RNN). When the observation signal is composed of the sum of the noise with the desired signal, it is a fundamental and important problem to reduce the noise in speech recognition, etc.. There is the noise canceller as the tool which removes the noise. The conventional noise canceller needs the reference input terminal. In this paper, the technique by which noise reduction is possible using RNN and the observation signal only, is proposed. Finally, the results of computer simulation are presented to illustrate the effectiveness of the proposed method.
收起
摘要 :
Recurrent neural network (RNN) and convolutional neural network (CNN) are two prevailing architectures used in text classification. Traditional approaches combine the strengths of these two networks by straightly streamlining them...
展开
Recurrent neural network (RNN) and convolutional neural network (CNN) are two prevailing architectures used in text classification. Traditional approaches combine the strengths of these two networks by straightly streamlining them or linking features extracted from them. In this article, a novel approach is proposed to maintain the strengths of RNN and CNN to a great extent. In the proposed approach, a bi-directional RNN encodes each word into forward and backward hidden states. Then, a neural tensor layer is used to fuse bi-directional hidden states to get word representations. Meanwhile, a convolutional neural network is utilized to learn the importance of each word for text classification. Empirical experiments are conducted on several datasets for text classification. The superior performance of the proposed approach confirms its effectiveness.
收起
摘要 :
Emotions play a vital role in the efficient and natural human computer interaction. Recognizing human emotions from their speech is truly a challenging task when accuracy, robustness and latency are considered. With the recent adv...
展开
Emotions play a vital role in the efficient and natural human computer interaction. Recognizing human emotions from their speech is truly a challenging task when accuracy, robustness and latency are considered. With the recent advancements in deep learning now it is possible to get better accuracy, robustness and low latency for solving complex functions. In our experiment we have developed two deep learning models for emotion recognition from speech. We compare the performance of a feed forward Deep Neural Network (DNN) with the recently developed Recurrent Neural Network (RNN) which is known as Gated Recurrent Unit (GRU) for speech emotion recognition. GRUs are currently not explored for classifying emotions from speech. The DNN model gives an accuracy of 89.96% and the GRU model gives an accuracy of 95.82%. Our experiments show that GRU model performs very well on emotion classification compared to the DNN model.
收起
摘要 :
This paper describes performance of different classifiers (established/combinations/new prediction methods) that are used in predicting stock price. Artificial Neural Network (ANN) was chosen as the target candidates for the forec...
展开
This paper describes performance of different classifiers (established/combinations/new prediction methods) that are used in predicting stock price. Artificial Neural Network (ANN) was chosen as the target candidates for the forecasting model in this work because of its ability to solvecomplex problems such as the stock price prediction. We experimented three types of neural network namely Feed Forward Neural Network (FFNN), Elman Recurrent Neural Network (ERNN), Jordan Recurrent Neural Network (JRNN) and compared their predictions’ accuracy. We then designed an ensembleneural network that combined FFNN, JRNN and ERNN using bagging method to build a more accurate predictive model. Based on the results obtained, our proposed ENN outperformed the other ANNs by achieving the highest prediction’s accuracy.
收起
摘要 :
Text is a significant tool for human communication, and text recognition in scene images becomes more and more important. In this paper, we propose a residual convolutional recurrent neural network for solving the task of scene te...
展开
Text is a significant tool for human communication, and text recognition in scene images becomes more and more important. In this paper, we propose a residual convolutional recurrent neural network for solving the task of scene text recognition. The general convolutional recurrent neural network (CRNN) is realized by combining convolutional neural network (CNN) with recurrent neural network (RNN). The CNN part extracts features and the RNN part encodes and decodes feature sequences. In order to improve the accuracy rate of scene text recognition based on CRNN, we explore different deeper CNN architectures to get feature descriptors and analyze the corresponding text recognition results. Specifically, VGG and ResNet are introduced to train these different deep models and obtain the encoding information of images. The experimental results on public datasets demonstrate the effectiveness of our method.
收起
摘要 :
It has been shown that neural networks are able to infer regular crisp grammars from positive and negative examples. The fuzzy grammatical inference (FGI) problem however has received considerably less attention. In this paper we ...
展开
It has been shown that neural networks are able to infer regular crisp grammars from positive and negative examples. The fuzzy grammatical inference (FGI) problem however has received considerably less attention. In this paper we show that a suitable two-layer neural network model is able to infer fuzzy regular grammars from a set of fuzzy examples belonging to a fuzzy language.
收起
摘要 :
Smart grid and microgrid technology based on energy storage systems (ESS) and renewable energy are attracting significant attention in addressing the challenges associated with climate change and energy crises. In particular, buil...
展开
Smart grid and microgrid technology based on energy storage systems (ESS) and renewable energy are attracting significant attention in addressing the challenges associated with climate change and energy crises. In particular, building an accurate short-term load forecasting (STLF) model for energy management systems (EMS) is a key factor in the successful formulation of an appropriate energy management strategy. Recent recurrent neural network (RNN)-based models have demonstrated favorable performance in electric load forecasting. However, when forecasting electric load at a specific time, existing RNN-based forecasting models neither use a predicted future hidden state vector nor the fully available past information. Therefore, once a hidden state vector has been incorrectly generated at a specific prediction time, it cannot be corrected for enhanced forecasting of the following prediction times. To address these problems, we propose a recurrent inception convolution neural network (RICNN) that combines RNN and 1-dimensional CNN (1-D CNN). We use the 1-D convolution inception module to calibrate the prediction time and the hidden state vector values calculated from nearby time steps. By doing so, the inception module generates an optimized network via the prediction time generated in the RNN and the nearby hidden state vectors. The proposed RICNN model has been verified in terms of the power usage data of three large distribution complexes in South Korea. Experimental results demonstrate that the RICNN model outperforms the benchmarked multi-layer perception, RNN, and 1-D CNN in daily electric load forecasting (48-time steps with an interval of 30 min). (C) 2019 Elsevier B.V. All rights reserved.
收起
原文获取