Visual metric and semantic localization for UGV by Zhang, Handuo

“…The fourth part of the thesis leverages the emerging feature learning framework. It proposes a unified self-supervised model called LGDNet to generate both global and local image feature descriptors end-to-end. …”
Get full text

Enhancing convolutional neural network predictions of electrocardiograms with left ventricular dysfunction using a novel sub-waveform representation by Hossein Honarvar, PhD, Chirag Agarwal, PhD, Sulaiman Somani, MD, Akhil Vaid, MD, Joshua Lampert, MD, Tingyi Wanyan, PhD, Vivek Y. Reddy, MD, Girish N. Nadkarni, MD, Riccardo Miotto, PhD, Marinka Zitnik, PhD, Fei Wang, PhD, Benjamin S. Glicksberg, PhD

“…In ECG DL, researchers often use convolutional neural networks (CNNs) and traditionally use the full duration of raw ECG waveforms that create redundancies in feature learning and result in inaccurate predictions with large uncertainties. …”
Get full text

Meta-analysis of Artificial Intelligence-Assisted Pathology for the Detection of Early Cervical Cancer by Di Qin, Chunmei Zhang, Huan Zhou, Xiaohui Yin, Geng Rong, Shixian Zhou, Mingming Wang, Zhigang Pei

“…The labeling method, image pre-processing method, and feature learning method affected the accuracy of the AI model in diagnosing early cervical cancer, while the choice of learning image library and validation database did not usually affect the accuracy of the model.…”
Get full text

A two-branch trade-off neural network for balanced scoring sleep stages on multiple cohorts by Di Zhang, Di Zhang, Jinbo Sun, Jinbo Sun, Yichong She, Yichong She, Yapeng Cui, Yapeng Cui, Xiao Zeng, Xiao Zeng, Liming Lu, Chunzhi Tang, Nenggui Xu, Badong Chen, Wei Qin, Wei Qin

“…A transitive training strategy is employed to balance universal feature learning and contextual referencing. Parameter optimization and benchmark comparisons are conducted using a large-scale dataset, followed by evaluation on seven datasets in five cohorts.ResultsThe proposed model achieves an accuracy of 88.16%, Cohen’s kappa of 0.836, and MF1 score of 0.818 on the SHHS1 test set, also with comparable performance to human scorers in scoring stage N1. …”
Get full text

Vision-based road lane and curb detection for unmanned ground vehicles in urban scenarios by Zhou, Hui

“…The third contribution of the thesis is to propose a unified network to incorporate these two tasks together by taking advantage of the powerful feature learning ability brought by deep convolutional neural networks. …”
Get full text

Development of lung segmentation method in x-ray images of children based on TransResUNet by Lingdong Chen, Lingdong Chen, Lingdong Chen, Zhuo Yu, Jian Huang, Jian Huang, Jian Huang, Liqi Shu, Pekka Kuosmanen, Pekka Kuosmanen, Chen Shen, Chen Shen, Chen Shen, Xiaohui Ma, Jing Li, Jing Li, Jing Li, Chensheng Sun, Chensheng Sun, Chensheng Sun, Zheming Li, Zheming Li, Zheming Li, Ting Shu, Gang Yu, Gang Yu, Gang Yu, Gang Yu

“…The former can provide a self-attention module improving the feature learning ability of the model, while the latter can avoid the problem of network degradation.ResultsApplied on the test set containing multi-center data, our model achieved a Dice score of 0.9822.ConclusionsThis novel lung segmentation method proposed in this work based on TransResUNet is better than other existing medical image segmentation networks.…”
Get full text

A cyber-physical system-enabled deep learning-based methodology for cognitive intelligent condition monitoring of manufacturing thing by Liu, Bufan

“…(Chapter 3) Secondly, an adaptive parallel feature learning and fusion approach is proposed to deal with the input of multiple data domains, feature importance weight generation, and handcrafted statistical feature fusion. …”
Get full text

Lightweight Intelligent Recognition of Saposhnikovia Divaricata (Turcz.) Schischk Originality Based on Improved ShuffleNet V2 by ZHAO Yu, REN Yiping, PIAO Xinru, ZHENG Danyang, LI Dongming

“…The SiLU activation function was used to replace the ReLU activation function to enhance the generalization ability of the model Enriching local feature learning. Therefore, a lightweight Shuffle-Hourglass SE model was proposed. …”
Get full text

Web text-aided image classification by Wang, Dongzhe

“…These proposed text feature learning methods are motivated by the transferable mid-level image representation learned by the convolutional neural network (CNN). …”
Get full text
Get full text

Representation learning with efficient extreme learning machine auto-encoders by Zhang, Guanghao

“…Extreme Learning Machine Auto-Encoder (ELM-AE) was proposed as a variant of general ELM for unsupervised feature learning. Specifically, ELM-AE introduces linear ELM-AE, nonlinear ELM-AE, and sparse ELM-AE. …”
Get full text

Learning visual representations without human supervision by Xie, Jiahao

“…We propose Online Deep Clustering (ODC) to tackle the long-standing challenge in traditional joint clustering and feature learning methods: the training schedule alternating between deep feature clustering and network parameters update leads to unstable learning of visual representations. …”
Get full text

Accident risk assessment and prediction using surrogate indicators and machine learning by Shi, Xiupeng

“…Furthermore, the linkages of risk levels and driving behaviours are explored, which empower behaviour-based risk prediction. An integrated feature learning framework is designed, to assess vehicle driving and predict risk levels. …”
Get full text

Person re-identification using visual analytic and deep learning by Tay, Chiat Pin

“…We also generate various heatmaps from the person attribute and combine them into an attention map at global level to further enhance the feature learning to get a stronger representation. With the learning of the attributes, we can perform additional query to the gallery images. …”
Get full text

Towards unbiased visual language reasoning and consistent segmentation by Huang, Jianqiang

“…Given a set of detected ob- ject regions in an image (e.g., by Faster R-CNN), like any other unsupervised feature learning methods (e.g., word2vec), the proxy training objective of VC R-CNN is to predict the contextual objects of a region. …”
Get full text

Deep learning for texture recognition: from streamlined architecture to multimodal extensions by Mao, Shangbo

“…The residual encoder is designed to preserve spatial information, thereby enhancing the feature learning process. In contrast, the aggregation module uses a simple averaging technique to generate spatially order-less features, improving the overall texture classification performance. …”
Get full text

Statistical analysis and machine learning in multimodal brain imaging of neuropsychiatric disorders by Hu, Mengjiao

“…The ability of automatic feature learning makes CNN potentially suitable to uncover the complex and widespread brain changes in schizophrenia. …”
Get full text

Entrepreneurship in Tourism / by Walmsley, Andreas, author

“…Structured into twelve chapters, this book takes an intuitive step-by step progression through entrepreneurship in tourism: context, theoretical perspectives and definitions; the entrepreneurial process from concept to reality; the business plan; marketing, employment and strategy to growth, policy context and future directions. Featuring learning outcomes, 'reflective practice' activities and a range of international case studies that encourage critical thinking and practical applications, this is essential reading for anyone studying tourism degree programmes at undergraduate and graduate level"…”

PRINCIPLES OF ANDRAGOGY IN MEDICAL EDUCATION by M. G. Romantsov, O. V. Shamsheva, I. Yu. Melnikova, A. A. Shuldyakov

“…The development of the modern education system includes a number of features. Learning involves a high level of independence with the leading role of the student. …”
Get full text

Explaining and Visualizing Embeddings of One-Dimensional Convolutional Models in Human Activity Recognition Tasks by Gustavo Aquino, Marly Guimarães Fernandes Costa, Cícero Ferreira Fernandes Costa Filho

“…However, understanding and explaining the features learned by these networks remains a challenge. …”
Get full text

Structured sparse K -means clustering via Laplacian smoothing by Gong, W, Zhao, R, Grünewald, S

“…At the same time, the relevant features learned by our method are more structured, hence have better interpretability. …”