Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
4.7
2.7
Journals
Entropy
Special Issues
Methods in Artificial Intelligence and Information Processing
entropy-logo
Submit to Entropy Review for Entropy Propose a Special Issue

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Methods in Artificial Intelligence and Information Processing

A special issue of Entropy (ISSN 1099-4300). This special issue belongs to the section "Signal and Data Analysis".

Deadline for manuscript submissions: closed (10 May 2022) | Viewed by 48977

Special Issue Editors

Prof. Dr. Zoran H. Perić

E-Mail Website
Guest Editor
Faculty of Electronic Engineering, University of Nis, 18106 Nis, Serbia
Interests: digital telecommunication; quantization; compression; machine learning; coding
Special Issues, Collections and Topics in MDPI journals
Prof. Dr. Vlado Delić

E-Mail Website
Guest Editor
Faculty of Technical Sciences, University of Novi Sad, 21102 Novi Sad, Serbia
Interests: ICT; speech technologies; HCI
Special Issues, Collections and Topics in MDPI journals
Dr. Vladimir Despotovic

E-Mail Website
Guest Editor
Faculty of Science, Technology and Communication, University of Luxembourg, 4365 Esch-sur-Alzette, Luxembourg
Interests: spoken language understanding; speech processing; machine learning; natural language processing; fractional calculus
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

The area of artificial intelligence (AI), although introduced many years ago, has received considerable attention nowadays. This can be explained by the necessity to process a large amount of data, where efficient methods and algorithms are desirable. Most of the AI methods encountered in the literature are based on the mathematical theory developed before occurring AI. Further research in this area will result in better understanding the AI, and will provide its simplification with corresponding approximations. Namely, such a simplification will provide the base for practical implementation, which is of crucial interest for engineers, researchers and scientists dealing with the transfer of scientific research results into commercial products and the other applications. On the other hand, designing and analyzing the processing algorithms by using only very complex mathematical theory in AI and information processing (IP), would result in a loss of wide applicability (professional and academic communities as well as possibility of hardware implementation).

Modern technology relies on research in IP and AI, and a number of methods have been developed with the aim of solving problems in: recognition and classification of signals (image, speech, audio, medical signals), recognition of emotions, signal quality enhancement, detection of signals in the presence of noise, pattern recognition in signals (speech, image, audio, biomedical signals), automatic diagnosis, methods and algorithms in wireless sensors networks, deep neural networks (DNN), data compression, data clustering, quantization in neural networks (NN) and learning representation.

This Special Issue concerns not only the application of methods but the promotion of the development in these two fields, independently and combined.

Potential topics include, but are not limited to, the following:

  • Parametric estimation in machine learning algorithms
  • Entropy and quantization
  • Entropy coding of signals
  • Entropy coding of data and parameters
  • Deep learning methods
  • Regression methods
  • Classification methods
  • Clustering methods
  • Neural networks (DNN, CNN,…)
  • Quantization methods in neural networks
  • Compression methods for neural networks and signal processing
  • Speech and image processing
  • Object detection and face recognition
  • Linear and non-linear prediction in signals and time series
  • Methods and algorithms for recognition and disease diagnosis in biomedical signals

Dr. Zoran H. Peric
Dr. Vlado Delic
Dr. Vladimir Despotovic
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Entropy is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • entropy and quantization
  • entropy coding of data and parameters
  • deep learning methods
  • cross entropy, classification methods
  • quantization methods in neural networks
  • speech and image processing
  • neural networks
  • prediction
  • compression methods

Related Special Issue

Published Papers (16 papers)

Download All Papers
Order results
Result details
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:

Research

Jump to: Review

16 pages, 3406 KiB  
Article
A Hybrid Deep Learning Model for Brain Tumour Classification
by Mohammed Rasool, Nor Azman Ismail, Wadii Boulila, Adel Ammar, Hussein Samma, Wael M. S. Yafooz and Abdel-Hamid M. Emara
Entropy 2022, 24(6), 799; https://doi.org/10.3390/e24060799 - 08 Jun 2022
Cited by 55 | Viewed by 5071
Abstract
A brain tumour is one of the major reasons for death in humans, and it is the tenth most common type of tumour that affects people of all ages. However, if detected early, it is one of the most treatable types of tumours. [...] Read more.
A brain tumour is one of the major reasons for death in humans, and it is the tenth most common type of tumour that affects people of all ages. However, if detected early, it is one of the most treatable types of tumours. Brain tumours are classified using biopsy, which is not usually performed before definitive brain surgery. An image classification technique for tumour diseases is important for accelerating the treatment process and avoiding surgery and errors from manual diagnosis by radiologists. The advancement of technology and machine learning (ML) can assist radiologists in tumour diagnostics using magnetic resonance imaging (MRI) images without invasive procedures. This work introduced a new hybrid CNN-based architecture to classify three brain tumour types through MRI images. The method suggested in this paper uses hybrid deep learning classification based on CNN with two methods. The first method combines a pre-trained Google-Net model of the CNN algorithm for feature extraction with SVM for pattern classification. The second method integrates a finely tuned Google-Net with a soft-max classifier. The proposed approach was evaluated using MRI brain images that contain a total of 1426 glioma images, 708 meningioma images, 930 pituitary tumour images, and 396 normal brain images. The reported results showed that an accuracy of 93.1% was achieved from the finely tuned Google-Net model. However, the synergy of Google-Net as a feature extractor with an SVM classifier improved recognition accuracy to 98.1%. Full article
(This article belongs to the Special Issue Methods in Artificial Intelligence and Information Processing)
►▼ Show Figures

Figure 1

19 pages, 991 KiB  
Article
Inter- and Intra-Modal Contrastive Hybrid Learning Framework for Multimodal Abstractive Summarization
by Jiangfeng Li, Zijian Zhang, Bowen Wang, Qinpei Zhao and Chenxi Zhang
Entropy 2022, 24(6), 764; https://doi.org/10.3390/e24060764 - 29 May 2022
Cited by 5 | Viewed by 1926
Abstract
Internet users are benefiting from technologies of abstractive summarization enabling them to view articles on the internet by reading article summaries only instead of an entire article. However, there are disadvantages to technologies for analyzing articles with texts and images due to the [...] Read more.
Internet users are benefiting from technologies of abstractive summarization enabling them to view articles on the internet by reading article summaries only instead of an entire article. However, there are disadvantages to technologies for analyzing articles with texts and images due to the semantic gap between vision and language. These technologies focus more on aggregating features and neglect the heterogeneity of each modality. At the same time, the lack of consideration of intrinsic data properties within each modality and semantic information from cross-modal correlations result in the poor quality of learned representations. Therefore, we propose a novel Inter- and Intra-modal Contrastive Hybrid learning framework which learns to automatically align the multimodal information and maintains the semantic consistency of input/output flows. Moreover, ITCH can be taken as a component to make the model suitable for both supervised and unsupervised learning approaches. Experiments on two public datasets, MMS and MSMO, show that the ITCH performances are better than the current baselines. Full article
(This article belongs to the Special Issue Methods in Artificial Intelligence and Information Processing)
►▼ Show Figures

Figure 1

17 pages, 2581 KiB  
Article
Speaker Recognition Using Constrained Convolutional Neural Networks in Emotional Speech
by Nikola Simić, Siniša Suzić, Tijana Nosek, Mia Vujović, Zoran Perić, Milan Savić and Vlado Delić
Entropy 2022, 24(3), 414; https://doi.org/10.3390/e24030414 - 16 Mar 2022
Cited by 9 | Viewed by 2543
Abstract
Speaker recognition is an important classification task, which can be solved using several approaches. Although building a speaker recognition model on a closed set of speakers under neutral speaking conditions is a well-researched task and there are solutions that provide excellent performance, the [...] Read more.
Speaker recognition is an important classification task, which can be solved using several approaches. Although building a speaker recognition model on a closed set of speakers under neutral speaking conditions is a well-researched task and there are solutions that provide excellent performance, the classification accuracy of developed models significantly decreases when applying them to emotional speech or in the presence of interference. Furthermore, deep models may require a large number of parameters, so constrained solutions are desirable in order to implement them on edge devices in the Internet of Things systems for real-time detection. The aim of this paper is to propose a simple and constrained convolutional neural network for speaker recognition tasks and to examine its robustness for recognition in emotional speech conditions. We examine three quantization methods for developing a constrained network: floating-point eight format, ternary scalar quantization, and binary scalar quantization. The results are demonstrated on the recently recorded SEAC dataset. Full article
(This article belongs to the Special Issue Methods in Artificial Intelligence and Information Processing)
►▼ Show Figures

Figure 1

16 pages, 3313 KiB  
Article
Deep Prediction Model Based on Dual Decomposition with Entropy and Frequency Statistics for Nonstationary Time Series
by Zhigang Shi, Yuting Bai, Xuebo Jin, Xiaoyi Wang, Tingli Su and Jianlei Kong
Entropy 2022, 24(3), 360; https://doi.org/10.3390/e24030360 - 02 Mar 2022
Cited by 9 | Viewed by 2252
Abstract
The prediction of time series is of great significance for rational planning and risk prevention. However, time series data in various natural and artificial systems are nonstationary and complex, which makes them difficult to predict. An improved deep prediction method is proposed herein [...] Read more.
The prediction of time series is of great significance for rational planning and risk prevention. However, time series data in various natural and artificial systems are nonstationary and complex, which makes them difficult to predict. An improved deep prediction method is proposed herein based on the dual variational mode decomposition of a nonstationary time series. First, criteria were determined based on information entropy and frequency statistics to determine the quantity of components in the variational mode decomposition, including the number of subsequences and the conditions for dual decomposition. Second, a deep prediction model was built for the subsequences obtained after the dual decomposition. Third, a general framework was proposed to integrate the data decomposition and deep prediction models. The method was verified on practical time series data with some contrast methods. The results show that it performed better than single deep network and traditional decomposition methods. The proposed method can effectively extract the characteristics of a nonstationary time series and obtain reliable prediction results. Full article
(This article belongs to the Special Issue Methods in Artificial Intelligence and Information Processing)
►▼ Show Figures

Figure 1

25 pages, 5419 KiB  
Article
Applying Hybrid ARIMA-SGARCH in Algorithmic Investment Strategies on S&P500 Index
by Nguyen Vo and Robert Ślepaczuk
Entropy 2022, 24(2), 158; https://doi.org/10.3390/e24020158 - 20 Jan 2022
Cited by 15 | Viewed by 4576
Abstract
This research aims to compare the performance of ARIMA as a linear model with that of the combination of ARIMA and GARCH family models to forecast S&P500 log returns in order to construct algorithmic investment strategies on this index. We used the data [...] Read more.
This research aims to compare the performance of ARIMA as a linear model with that of the combination of ARIMA and GARCH family models to forecast S&P500 log returns in order to construct algorithmic investment strategies on this index. We used the data collected from Yahoo Finance with daily frequency for the period from 1 January 2000 to 31 December 2019. By using a rolling window approach, we compared ARIMA with the hybrid models to examine whether hybrid ARIMA-SGARCH and ARIMA-EGARCH can really reflect the specific time-series characteristics and have better predictive power than the simple ARIMA model. In order to assess the precision and quality of these models in forecasting, we compared their equity lines, their forecasting error metrics (MAE, MAPE, RMSE, MAPE), and their performance metrics (annualized return compounded, annualized standard deviation, maximum drawdown, information ratio, and adjusted information ratio). The main contribution of this research is to show that the hybrid models outperform ARIMA and the benchmark (Buy&Hold strategy on S&P500 index) over the long term. These results are not sensitive to varying window sizes, the type of distribution, and the type of the GARCH model. Full article
(This article belongs to the Special Issue Methods in Artificial Intelligence and Information Processing)
►▼ Show Figures

Figure 1

15 pages, 1956 KiB  
Article
Cooperative Spectrum Sensing Based on Multi-Features Combination Network in Cognitive Radio Network
by Mingdong Xu, Zhendong Yin, Yanlong Zhao and Zhilu Wu
Entropy 2022, 24(1), 129; https://doi.org/10.3390/e24010129 - 15 Jan 2022
Cited by 12 | Viewed by 2345
Abstract
Cognitive radio, as a key technology to improve the utilization of radio spectrum, acquired much attention. Moreover, spectrum sensing has an irreplaceable position in the field of cognitive radio and was widely studied. The convolutional neural networks (CNNs) and the gate recurrent unit [...] Read more.
Cognitive radio, as a key technology to improve the utilization of radio spectrum, acquired much attention. Moreover, spectrum sensing has an irreplaceable position in the field of cognitive radio and was widely studied. The convolutional neural networks (CNNs) and the gate recurrent unit (GRU) are complementary in their modelling capabilities. In this paper, we introduce a CNN-GRU network to obtain the local information for single-node spectrum sensing, in which CNN is used to extract spatial feature and GRU is used to extract the temporal feature. Then, the combination network receives the features extracted by the CNN-GRU network to achieve multifeatures combination and obtains the final cooperation result. The cooperative spectrum sensing scheme based on Multifeatures Combination Network enhances the sensing reliability by fusing the local information from different sensing nodes. To accommodate the detection of multiple types of signals, we generated 8 kinds of modulation types to train the model. Theoretical analysis and simulation results show that the cooperative spectrum sensing algorithm proposed in this paper improved detection performance with no prior knowledge about the information of primary user or channel state. Our proposed method achieved competitive performance under the condition of large dynamic signal-to-noise ratio. Full article
(This article belongs to the Special Issue Methods in Artificial Intelligence and Information Processing)
►▼ Show Figures

Figure 1

18 pages, 5346 KiB  
Article
Evaluating the Learning Procedure of CNNs through a Sequence of Prognostic Tests Utilising Information Theoretical Measures
by Xiyu Shi, Varuna De-Silva, Yusuf Aslan, Erhan Ekmekcioglu and Ahmet Kondoz
Entropy 2022, 24(1), 67; https://doi.org/10.3390/e24010067 - 30 Dec 2021
Cited by 2 | Viewed by 2395
Abstract
Deep learning has proven to be an important element of modern data processing technology, which has found its application in many areas such as multimodal sensor data processing and understanding, data generation and anomaly detection. While the use of deep learning is booming [...] Read more.
Deep learning has proven to be an important element of modern data processing technology, which has found its application in many areas such as multimodal sensor data processing and understanding, data generation and anomaly detection. While the use of deep learning is booming in many real-world tasks, the internal processes of how it draws results is still uncertain. Understanding the data processing pathways within a deep neural network is important for transparency and better resource utilisation. In this paper, a method utilising information theoretic measures is used to reveal the typical learning patterns of convolutional neural networks, which are commonly used for image processing tasks. For this purpose, training samples, true labels and estimated labels are considered to be random variables. The mutual information and conditional entropy between these variables are then studied using information theoretical measures. This paper shows that more convolutional layers in the network improve its learning and unnecessarily higher numbers of convolutional layers do not improve the learning any further. The number of convolutional layers that need to be added to a neural network to gain the desired learning level can be determined with the help of theoretic information quantities including entropy, inequality and mutual information among the inputs to the network. The kernel size of convolutional layers only affects the learning speed of the network. This study also shows that where the dropout layer is applied to has no significant effects on the learning of networks with a lower dropout rate, and it is better placed immediately after the last convolutional layer with higher dropout rates. Full article
(This article belongs to the Special Issue Methods in Artificial Intelligence and Information Processing)
►▼ Show Figures

Figure 1

19 pages, 1495 KiB  
Article
Artificial Neural Networks Performance in WIG20 Index Options Pricing
by Maciej Wysocki and Robert Ślepaczuk
Entropy 2022, 24(1), 35; https://doi.org/10.3390/e24010035 - 24 Dec 2021
Cited by 7 | Viewed by 2830
Abstract
In this paper, the performance of artificial neural networks in option pricing was analyzed and compared with the results obtained from the Black–Scholes–Merton model, based on the historical volatility. The results were compared based on various error metrics calculated separately between three moneyness [...] Read more.
In this paper, the performance of artificial neural networks in option pricing was analyzed and compared with the results obtained from the Black–Scholes–Merton model, based on the historical volatility. The results were compared based on various error metrics calculated separately between three moneyness ratios. The market data-driven approach was taken to train and test the neural network on the real-world options data from 2009 to 2019, quoted on the Warsaw Stock Exchange. The artificial neural network did not provide more accurate option prices, even though its hyperparameters were properly tuned. The Black–Scholes–Merton model turned out to be more precise and robust to various market conditions. In addition, the bias of the forecasts obtained from the neural network differed significantly between moneyness states. This study provides an initial insight into the application of deep learning methods to pricing options in emerging markets with low liquidity and high volatility. Full article
(This article belongs to the Special Issue Methods in Artificial Intelligence and Information Processing)
►▼ Show Figures

Figure 1

28 pages, 7757 KiB  
Article
Whether the Support Region of Three-Bit Uniform Quantizer Has a Strong Impact on Post-Training Quantization for MNIST Dataset?
by Jelena Nikolić, Zoran Perić, Danijela Aleksić, Stefan Tomić and Aleksandra Jovanović
Entropy 2021, 23(12), 1699; https://doi.org/10.3390/e23121699 - 20 Dec 2021
Cited by 6 | Viewed by 2691
Abstract
Driven by the need for the compression of weights in neural networks (NNs), which is especially beneficial for edge devices with a constrained resource, and by the need to utilize the simplest possible quantization model, in this paper, we study the performance of [...] Read more.
Driven by the need for the compression of weights in neural networks (NNs), which is especially beneficial for edge devices with a constrained resource, and by the need to utilize the simplest possible quantization model, in this paper, we study the performance of three-bit post-training uniform quantization. The goal is to put various choices of the key parameter of the quantizer in question (support region threshold) in one place and provide a detailed overview of this choice’s impact on the performance of post-training quantization for the MNIST dataset. Specifically, we analyze whether it is possible to preserve the accuracy of the two NN models (MLP and CNN) to a great extent with the very simple three-bit uniform quantizer, regardless of the choice of the key parameter. Moreover, our goal is to answer the question of whether it is of the utmost importance in post-training three-bit uniform quantization, as it is in quantization, to determine the optimal support region threshold value of the quantizer to achieve some predefined accuracy of the quantized neural network (QNN). The results show that the choice of the support region threshold value of the three-bit uniform quantizer does not have such a strong impact on the accuracy of the QNNs, which is not the case with two-bit uniform post-training quantization, when applied in MLP for the same classification task. Accordingly, one can anticipate that due to this special property, the post-training quantization model in question can be greatly exploited. Full article
(This article belongs to the Special Issue Methods in Artificial Intelligence and Information Processing)
►▼ Show Figures

Figure 1

15 pages, 2565 KiB  
Article
An Improved K-Means Algorithm Based on Evidence Distance
by Ailin Zhu, Zexi Hua, Yu Shi, Yongchuan Tang and Lingwei Miao
Entropy 2021, 23(11), 1550; https://doi.org/10.3390/e23111550 - 21 Nov 2021
Cited by 9 | Viewed by 3597
Abstract
The main influencing factors of the clustering effect of the k-means algorithm are the selection of the initial clustering center and the distance measurement between the sample points. The traditional k-mean algorithm uses Euclidean distance to measure the distance between sample points, thus [...] Read more.
The main influencing factors of the clustering effect of the k-means algorithm are the selection of the initial clustering center and the distance measurement between the sample points. The traditional k-mean algorithm uses Euclidean distance to measure the distance between sample points, thus it suffers from low differentiation of attributes between sample points and is prone to local optimal solutions. For this feature, this paper proposes an improved k-means algorithm based on evidence distance. Firstly, the attribute values of sample points are modelled as the basic probability assignment (BPA) of sample points. Then, the traditional Euclidean distance is replaced by the evidence distance for measuring the distance between sample points, and finally k-means clustering is carried out using UCI data. Experimental comparisons are made with the traditional k-means algorithm, the k-means algorithm based on the aggregation distance parameter, and the Gaussian mixture model. The experimental results show that the improved k-means algorithm based on evidence distance proposed in this paper has a better clustering effect and the convergence of the algorithm is also better. Full article
(This article belongs to the Special Issue Methods in Artificial Intelligence and Information Processing)
►▼ Show Figures

Figure 1

18 pages, 538 KiB  
Article
Mood Disorder Detection in Adolescents by Classification Trees, Random Forests and XGBoost in Presence of Missing Data
by Elzbieta Turska, Szymon Jurga and Jaroslaw Piskorski
Entropy 2021, 23(9), 1210; https://doi.org/10.3390/e23091210 - 14 Sep 2021
Cited by 8 | Viewed by 2249
Abstract
We apply tree-based classification algorithms, namely the classification trees, with the use of the rpart algorithm, random forests and XGBoost methods to detect mood disorder in a group of 2508 lower secondary school students. The dataset presents many challenges, the most important of [...] Read more.
We apply tree-based classification algorithms, namely the classification trees, with the use of the rpart algorithm, random forests and XGBoost methods to detect mood disorder in a group of 2508 lower secondary school students. The dataset presents many challenges, the most important of which is many missing data as well as the being heavily unbalanced (there are few severe mood disorder cases). We find that all algorithms are specific, but only the rpart algorithm is sensitive; i.e., it is able to detect cases of real cases mood disorder. The conclusion of this paper is that this is caused by the fact that the rpart algorithm uses the surrogate variables to handle missing data. The most important social-studies-related result is that the adolescents’ relationships with their parents are the single most important factor in developing mood disorders—far more important than other factors, such as the socio-economic status or school success. Full article
(This article belongs to the Special Issue Methods in Artificial Intelligence and Information Processing)
►▼ Show Figures

Figure 1

21 pages, 3991 KiB  
Article
Magnetic Resonance Imaging Segmentation via Weighted Level Set Model Based on Local Kernel Metric and Spatial Constraint
by Jianhua Song and Zhe Zhang
Entropy 2021, 23(9), 1196; https://doi.org/10.3390/e23091196 - 10 Sep 2021
Cited by 7 | Viewed by 1966
Abstract
Magnetic resonance imaging (MRI) segmentation is a fundamental and significant task since it can guide subsequent clinic diagnosis and treatment. However, images are often corrupted by defects such as low-contrast, noise, intensity inhomogeneity, and so on. Therefore, a weighted level set model (WLSM) [...] Read more.
Magnetic resonance imaging (MRI) segmentation is a fundamental and significant task since it can guide subsequent clinic diagnosis and treatment. However, images are often corrupted by defects such as low-contrast, noise, intensity inhomogeneity, and so on. Therefore, a weighted level set model (WLSM) is proposed in this study to segment inhomogeneous intensity MRI destroyed by noise and weak boundaries. First, in order to segment the intertwined regions of brain tissue accurately, a weighted neighborhood information measure scheme based on local multi information and kernel function is designed. Then, the membership function of fuzzy c-means clustering is used as the spatial constraint of level set model to overcome the sensitivity of level set to initialization, and the evolution of level set function can be adaptively changed according to different tissue information. Finally, the distance regularization term in level set function is replaced by a double potential function to ensure the stability of the energy function in the evolution process. Both real and synthetic MRI images can show the effectiveness and performance of WLSM. In addition, compared with several state-of-the-art models, segmentation accuracy and Jaccard similarity coefficient obtained by WLSM are increased by 0.0586, 0.0362 and 0.1087, 0.0703, respectively. Full article
(This article belongs to the Special Issue Methods in Artificial Intelligence and Information Processing)
►▼ Show Figures

Figure 1

17 pages, 5729 KiB  
Article
Design of a 2-Bit Neural Network Quantizer for Laplacian Source
by Zoran Perić, Milan Savić, Nikola Simić, Bojan Denić and Vladimir Despotović
Entropy 2021, 23(8), 933; https://doi.org/10.3390/e23080933 - 22 Jul 2021
Cited by 9 | Viewed by 2438
Abstract
Achieving real-time inference is one of the major issues in contemporary neural network applications, as complex algorithms are frequently being deployed to mobile devices that have constrained storage and computing power. Moving from a full-precision neural network model to a lower representation by [...] Read more.
Achieving real-time inference is one of the major issues in contemporary neural network applications, as complex algorithms are frequently being deployed to mobile devices that have constrained storage and computing power. Moving from a full-precision neural network model to a lower representation by applying quantization techniques is a popular approach to facilitate this issue. Here, we analyze in detail and design a 2-bit uniform quantization model for Laplacian source due to its significance in terms of implementation simplicity, which further leads to a shorter processing time and faster inference. The results show that it is possible to achieve high classification accuracy (more than 96% in the case of MLP and more than 98% in the case of CNN) by implementing the proposed model, which is competitive to the performance of the other quantization solutions with almost optimal precision. Full article
(This article belongs to the Special Issue Methods in Artificial Intelligence and Information Processing)
►▼ Show Figures

Figure 1

14 pages, 581 KiB  
Article
Augmenting Paraphrase Generation with Syntax Information Using Graph Convolutional Networks
by Xiaoqiang Chi and Yang Xiang
Entropy 2021, 23(5), 566; https://doi.org/10.3390/e23050566 - 02 May 2021
Cited by 4 | Viewed by 2573
Abstract
Paraphrase generation is an important yet challenging task in natural language processing. Neural network-based approaches have achieved remarkable success in sequence-to-sequence learning. Previous paraphrase generation work generally ignores syntactic information regardless of its availability, with the assumption that neural nets could learn such [...] Read more.
Paraphrase generation is an important yet challenging task in natural language processing. Neural network-based approaches have achieved remarkable success in sequence-to-sequence learning. Previous paraphrase generation work generally ignores syntactic information regardless of its availability, with the assumption that neural nets could learn such linguistic knowledge implicitly. In this work, we make an endeavor to probe into the efficacy of explicit syntactic information for the task of paraphrase generation. Syntactic information can appear in the form of dependency trees, which could be easily acquired from off-the-shelf syntactic parsers. Such tree structures could be conveniently encoded via graph convolutional networks to obtain more meaningful sentence representations, which could improve generated paraphrases. Through extensive experiments on four paraphrase datasets with different sizes and genres, we demonstrate the utility of syntactic information in neural paraphrase generation under the framework of sequence-to-sequence modeling. Specifically, our graph convolutional network-enhanced models consistently outperform their syntax-agnostic counterparts using multiple evaluation metrics. Full article
(This article belongs to the Special Issue Methods in Artificial Intelligence and Information Processing)
►▼ Show Figures

Figure 1

19 pages, 2349 KiB  
Article
Integrate Candidate Answer Extraction with Re-Ranking for Chinese Machine Reading Comprehension
by Junjie Zeng, Xiaoya Sun, Qi Zhang and Xinmeng Li
Entropy 2021, 23(3), 322; https://doi.org/10.3390/e23030322 - 08 Mar 2021
Viewed by 1827
Abstract
Machine Reading Comprehension (MRC) research concerns how to endow machines with the ability to understand given passages and answer questions, which is a challenging problem in the field of natural language processing. To solve the Chinese MRC task efficiently, this paper proposes an [...] Read more.
Machine Reading Comprehension (MRC) research concerns how to endow machines with the ability to understand given passages and answer questions, which is a challenging problem in the field of natural language processing. To solve the Chinese MRC task efficiently, this paper proposes an Improved Extraction-based Reading Comprehension method with Answer Re-ranking (IERC-AR), consisting of a candidate answer extraction module and a re-ranking module. The candidate answer extraction module uses an improved pre-training language model, RoBERTa-WWM, to generate precise word representations, which can solve the problem of polysemy and is good for capturing Chinese word-level features. The re-ranking module re-evaluates candidate answers based on a self-attention mechanism, which can improve the accuracy of predicting answers. Traditional machine-reading methods generally integrate different modules into a pipeline system, which leads to re-encoding problems and inconsistent data distribution between the training and testing phases; therefore, this paper proposes an end-to-end model architecture for IERC-AR to reasonably integrate the candidate answer extraction and re-ranking modules. The experimental results on the Les MMRC dataset show that IERC-AR outperforms state-of-the-art MRC approaches. Full article
(This article belongs to the Special Issue Methods in Artificial Intelligence and Information Processing)
►▼ Show Figures

Figure 1

Review

Jump to: Research

17 pages, 610 KiB  
Review
Singing Voice Detection: A Survey
by Ramy Monir, Daniel Kostrzewa and Dariusz Mrozek
Entropy 2022, 24(1), 114; https://doi.org/10.3390/e24010114 - 12 Jan 2022
Cited by 10 | Viewed by 3268
Abstract
Singing voice detection or vocal detection is a classification task that determines whether there is a singing voice in a given audio segment. This process is a crucial preprocessing step that can be used to improve the performance of other tasks such as [...] Read more.
Singing voice detection or vocal detection is a classification task that determines whether there is a singing voice in a given audio segment. This process is a crucial preprocessing step that can be used to improve the performance of other tasks such as automatic lyrics alignment, singing melody transcription, singing voice separation, vocal melody extraction, and many more. This paper presents a survey on the techniques of singing voice detection with a deep focus on state-of-the-art algorithms such as convolutional LSTM and GRU-RNN. It illustrates a comparison between existing methods for singing voice detection, mainly based on the Jamendo and RWC datasets. Long-term recurrent convolutional networks have reached impressive results on public datasets. The main goal of the present paper is to investigate both classical and state-of-the-art approaches to singing voice detection. Full article
(This article belongs to the Special Issue Methods in Artificial Intelligence and Information Processing)
►▼ Show Figures

Figure 1

Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-16
Back to TopTop