scholarly journals Gaussian Transformer: A Lightweight Approach for Natural Language Inference

2019 ◽  
Vol 33 ◽  
pp. 6489-6496 ◽  
Author(s):  
Maosheng Guo ◽  
Yu Zhang ◽  
Ting Liu
Keyword(s):  
Neural Networks ◽  
Natural Language ◽  
State Of The Art ◽  
Research Area ◽  
High Order Interaction ◽  
Training Time ◽  
Attention Networks ◽  
Local Dependency ◽  
Active Research ◽  
Active Research Area

Natural Language Inference (NLI) is an active research area, where numerous approaches based on recurrent neural networks (RNNs), convolutional neural networks (CNNs), and self-attention networks (SANs) has been proposed. Although obtaining impressive performance, previous recurrent approaches are hard to train in parallel; convolutional models tend to cost more parameters, while self-attention networks are not good at capturing local dependency of texts. To address this problem, we introduce a Gaussian prior to selfattention mechanism, for better modeling the local structure of sentences. Then we propose an efficient RNN/CNN-free architecture named Gaussian Transformer for NLI, which consists of encoding blocks modeling both local and global dependency, high-order interaction blocks collecting the evidence of multi-step inference, and a lightweight comparison block saving lots of parameters. Experiments show that our model achieves new state-of-the-art performance on both SNLI and MultiNLI benchmarks with significantly fewer parameters and considerably less training time. Besides, evaluation using the Hard NLI datasets demonstrates that our approach is less affected by the undesirable annotation artifacts.

scholarly journals Global-local attention for emotion recognition

2021 ◽  
Author(s):  
Nhat Le ◽  
Khanh Nguyen ◽  
Anh Nguyen ◽  
Bac Le
Keyword(s):  
Emotion Recognition ◽  
State Of The Art ◽  
Contextual Information ◽  
Research Area ◽  
The Past ◽  
Current State ◽  
Substantial Progress ◽  
Human Emotions ◽  
Active Research ◽  
Active Research Area

AbstractHuman emotion recognition is an active research area in artificial intelligence and has made substantial progress over the past few years. Many recent works mainly focus on facial regions to infer human affection, while the surrounding context information is not effectively utilized. In this paper, we proposed a new deep network to effectively recognize human emotions using a novel global-local attention mechanism. Our network is designed to extract features from both facial and context regions independently, then learn them together using the attention module. In this way, both the facial and contextual information is used to infer human emotions, therefore enhancing the discrimination of the classifier. The intensive experiments show that our method surpasses the current state-of-the-art methods on recent emotion datasets by a fair margin. Qualitatively, our global-local attention module can extract more meaningful attention maps than previous methods. The source code and trained model of our network are available at https://github.com/minhnhatvt/glamor-net.

Machine Translation

2016 ◽  
Author(s):  
Lucia Specia ◽  
Yorick Wilks
Keyword(s):  
Machine Translation ◽  
Language Processing ◽  
State Of The Art ◽  
Research Area ◽  
Rule Based ◽  
Active Research ◽  
Translation Methods ◽  
The Cost ◽  
Translation Systems ◽  
Active Research Area

Machine Translation (MT) is and always has been a core application in the field of natural-language processing. It is a very active research area and it has been attracting significant commercial interest, most of which has been driven by the deployment of corpus-based, statistical approaches, which can be built in a much shorter time and at a fraction of the cost of traditional, rule-based approaches, and yet produce translations of comparable or superior quality. This chapter aims at introducing MT and its main approaches. It provides a historical overview of the field, an introduction to different translation methods, both rationalist (rule-based) and empirical, and a more in depth description of state-of-the-art statistical methods. Finally, it covers popular metrics to evaluate the output of machine translation systems.

scholarly journals Leaf Features Extraction for Plant Classification using CNN

2021 ◽  
pp. 148-154
Author(s):  
Mr. P. Siva Prasad ◽  
Dr. A. Senthilrajan
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Neural Networks ◽  
Deep Learning ◽  
Deep Neural Network ◽  
Research Area ◽  
Network Approach ◽  
Neural Network Approach ◽  
Active Research ◽  
Active Research Area

Deep learning is now an active research area. Deep learning has done a success in computer vision and image recognition. It is a subset of the Machine Learning. In Deep learning, Convolutional Neural Network (CNN) is popular deep neural network approach. In this paper, we have addressed that how to extract useful leaf features automatically from the leaf dataset through Convolutional Neural Networks (CNN) using Deep Learning. In this paper, we have shown that the accuracy obtained by CNN approach is efficient when compared to accuracy obtained by the traditional neural network.

A Person-Centered Design Framework for Serious Games for Dementia

2020 ◽  
Vol 64 (1) ◽  
pp. 18-22
Author(s):  
Bella Yigong Zhang ◽  
Mark Chignell
Keyword(s):  
Theoretical Basis ◽  
Serious Games ◽  
Aging Population ◽  
Research Area ◽  
Design Framework ◽  
Strategic Design ◽  
Health And Wellbeing ◽  
Initial Design ◽  
Active Research ◽  
Active Research Area

With the rapidly aging population and the rising number of people living with dementia (PLWD), there is an urgent need for programming and activities that can promote the health and wellbeing of PLWD. Due to staffing and budgetary constraints, there is considerable interest in using technology to support this effort. Serious games for dementia have become a very active research area. However, much of the work is being done without a strong theoretical basis. We incorporate a Montessori approach with highly tactile interactions. We have developed a person-centered design framework for serious games for dementia with initial design recommendations. This framework has the potential to facilitate future strategic design and development in the field of serious games for dementia.

scholarly journals Digitization and Visualization of Folk Dances in Cultural Heritage: A Review

Inventions ◽  
2018 ◽  
Vol 3 (4) ◽  
pp. 72 ◽  
Author(s):  
Iris Kico ◽  
Nikos Grammalidis ◽  
Yiannis Christidis ◽  
Fotis Liarokapis
Keyword(s):  
Performance Evaluation ◽  
Cultural Heritage ◽  
Computer Science ◽  
Learning Process ◽  
Research Area ◽  
Active Research ◽  
Active Research Area

According to UNESCO, cultural heritage does not only include monuments and collections of objects, but also contains traditions or living expressions inherited from our ancestors and passed to our descendants. Folk dances represent part of cultural heritage and their preservation for the next generations appears of major importance. Digitization and visualization of folk dances form an increasingly active research area in computer science. In parallel to the rapidly advancing technologies, new ways for learning folk dances are explored, making the digitization and visualization of assorted folk dances for learning purposes using different equipment possible. Along with challenges and limitations, solutions that can assist the learning process and provide the user with meaningful feedback are proposed. In this paper, an overview of the techniques used for the recording of dance moves is presented. The different ways of visualization and giving the feedback to the user are reviewed as well as ways of performance evaluation. This paper reviews advances in digitization and visualization of folk dances from 2000 to 2018.

scholarly journals Enhancing Big Data Auditing

2018 ◽  
Vol 11 (1) ◽  
pp. 90
Author(s):  
Sara Alomari ◽  
Mona Alghamdi ◽  
Fahd S. Alotaibi
Keyword(s):  
Big Data ◽  
Research Area ◽  
Provable Data Possession ◽  
Integrity Verification ◽  
The Core ◽  
Outsourced Data ◽  
Active Research ◽  
Data Auditing ◽  
Active Research Area ◽  
Proof Of Retrievability

The auditing services of the outsourced data, especially big data, have been an active research area recently. Many schemes of remotely data auditing (RDA) have been proposed. Both categories of RDA, which are Provable Data Possession (PDP) and Proof of Retrievability (PoR), mostly represent the core schemes for most researchers to derive new schemes that support additional capabilities such as batch and dynamic auditing. In this paper, we choose the most popular PDP schemes to be investigated due to the existence of many PDP techniques which are further improved to achieve efficient integrity verification. We firstly review the work of literature to form the required knowledge about the auditing services and related schemes. Secondly, we specify a methodology to be adhered to attain the research goals. Then, we define each selected PDP scheme and the auditing properties to be used to compare between the chosen schemes. Therefore, we decide, if possible, which scheme is optimal in handling big data auditing.

scholarly journals E-Mail Usage in South Pacific Distance Education

2005 ◽  
pp. 1034-1039
Author(s):  
Jonathan Frank ◽  
Janet Toland ◽  
Karen D. Schenk
Keyword(s):  
Learning Styles ◽  
Research Area ◽  
Web Based ◽  
Group Interactions ◽  
Face To Face ◽  
Web Based Learning ◽  
Active Research ◽  
The Impact ◽  
E Mail ◽  
Active Research Area

The impact of cultural diversity on group interactions through technology is an active research area. Current research has found that a student’s culture appears to influence online interactions with teachers and other students (Freedman & Liu, 1996). Students from Asian and Western cultures have different Web-based learning styles (Liang & McQueen, 1999), and Scandinavian students demonstrate a more restrained online presence compared to their more expressive American counterparts (Bannon, 1995). Differences were also found across cultures in online compared to face-to-face discussions (Warschauer, 1996). Student engagement, discourse, and interaction are valued highly in “western” universities. With growing internationalization of western campuses, increasing use of educational technology both on and off campus, and rising distance learning enrollments, intercultural frictions are bound to increase.

scholarly journals Word and Chracter Segmentation in Devnagari and Odia Script – A Comparitive Analysis

2020 ◽  
Vol 9 (9) ◽  
pp. 377-382
Keyword(s):  
Character Recognition ◽  
Optical Character Recognition ◽  
Research Area ◽  
Character Segmentation ◽  
National Language ◽  
Eastern India ◽  
Regional Language ◽  
Optical Character ◽  
Active Research ◽  
Active Research Area

Optical Character Recognition has been an active research area in computer science for several years. Several research works undertaken on various languages in India. In this paper an attempt has been made to find out the percentage of accuracy in word and character segmentation of Hindi (National language of India) and Odia is one of the Regional Language mostly spoken in Odisha and a few Eastern India states. A comparative article has been published under this article. 10 sets of each printed Odia and Devanagari scripts with different word limits were used in this study. The documents were scanned at 300dpi before adopting pre-processing and segmentation procedure. The result shows that the percentage of accuracy both in word and character segmentation is higher in Odia language as compared to Hindi language. One of the reasons is the use of headers line in Hindi which makes the segmentation process cumbersome. Thus, it can be concluded that the accuracy level can vary from one language to the other and from word segmentation to that of the character segmentation.

scholarly journals Which OCR toolset is good and why? A comparative study

2021 ◽  
Vol 48 (2) ◽  
Author(s):  
Pooja Jain ◽  
◽  
Dr. Kavita Taneja ◽  
Dr. Harmunish Taneja ◽  
◽  
...  
Keyword(s):  
Comparative Study ◽  
Language Processing ◽  
Character Recognition ◽  
Optical Character Recognition ◽  
Research Area ◽  
Real World Applications ◽  
Banking Education ◽  
Active Research ◽  
Active Research Area ◽  
Computational Technology

Optical Character Recognition (OCR) is a very active research area in many challenging fields like pattern recognition, natural language processing (NLP), computer vision, biomedical informatics, machine learning (ML), and artificial intelligence (AI). This computational technology extracts the text in an editable format (MS Word/Excel, text files, etc.) from PDF files, scanned or hand-written documents, images (photographs, advertisements, and alike), etc. for further processing and has been utilized in many real-world applications including banking, education, insurance, finance, healthcare and keyword-based search in documents, etc. Many OCR toolsets are available under various categories, including open-source, proprietary, and online services. This research paper provides a comparative study of various OCR toolsets considering a variety of parameters.

scholarly journals Soft Threshold Ternary Networks

2020 ◽  
Author(s):  
Weixiang Xu ◽  
Xiangyu He ◽  
Tianli Zhao ◽  
Qinghao Hu ◽  
Peisong Wang ◽  
...  
Keyword(s):  
Neural Networks ◽  
Mobile Devices ◽  
State Of The Art ◽  
Performance Gap ◽  
Training Time ◽  
Current State ◽  
The Arts ◽  
Soft Threshold ◽  
And Storage ◽  
Selection Of

Large neural networks are difficult to deploy on mobile devices because of intensive computation and storage. To alleviate it, we study ternarization, a balance between efficiency and accuracy that quantizes both weights and activations into ternary values. In previous ternarized neural networks, a hard threshold Δ is introduced to determine quantization intervals. Although the selection of Δ greatly affects the training results, previous works estimate Δ via an approximation or treat it as a hyper-parameter, which is suboptimal. In this paper, we present the Soft Threshold Ternary Networks (STTN), which enables the model to automatically determine quantization intervals instead of depending on a hard threshold. Concretely, we replace the original ternary kernel with the addition of two binary kernels at training time, where ternary values are determined by the combination of two corresponding binary values. At inference time, we add up the two binary kernels to obtain a single ternary kernel. Our method dramatically outperforms current state-of-the-arts, lowering the performance gap between full-precision networks and extreme low bit networks. Experiments on ImageNet with AlexNet (Top-1 55.6%), ResNet-18 (Top-1 66.2%) achieves new state-of-the-art.

Sign in / Sign up

Export Citation Format

Share Document