☆ 4.7 Article

Self-Supervised Learning for data scarcity in a fatigue damage prognostic problem

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE (2023)

相关参考文献

注意：仅列出部分参考文献，下载原文获取全部文献信息。

Article Computer Science, Artificial Intelligence

Self-Supervised Learning: Generative or Contrastive

Xiao Liu et al.

Summary: Deep supervised learning has been successful, but it is limited by manual labels and vulnerable to attacks. In contrast, self-supervised learning utilizes input data as supervision, showing promising performance on representation learning. This survey comprehensively reviews self-supervised learning methods in computer vision, natural language processing, and graph learning.

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING (2023)

添加到收藏夹

Review Computer Science, Theory & Methods

A Systematic Review on Data Scarcity Problem in Deep Learning: Solution and Applications

Aayushi Bansal et al.

Summary: Recent advancements in deep learning architecture have made it more useful in real-life applications. However, training deep learning models requires a large amount of data, which may not be readily available in many domains. Data augmentation techniques can be used to address this problem by expanding the available dataset, improving model accuracy and generalization.

ACM COMPUTING SURVEYS (2022)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Deep Kronecker neural networks: A general framework for neural networks with adaptive activation functions

Ameya D. Jagtap et al.

Summary: Kronecker neural networks (KNNs) are a new type of neural networks with adaptive activation functions, utilizing the Kronecker product to construct wide networks with low parameters. Theoretical analysis and empirical examples show that KNNs have faster loss decay compared to feed-forward networks under certain conditions. Additionally, the proposed Rowdy activation function eliminates saturation regions and can be applied in various neural network architectures, demonstrating effectiveness in computational experiments.

NEUROCOMPUTING (2022)

添加到收藏夹

Article Engineering, Industrial

Self-supervised pretraining via contrast learning for intelligent incipient fault detection of bearings

Yifei Ding et al.

Summary: This paper introduces a self-supervised pre-training method via contrast learning (SSPCL) to learn discriminative representations from unlabeled bearing datasets. By proposing a specific architecture and data augmentation techniques, successfully applied to the deployment of models for bearing vibration signals, achieving early fault detection for the run-to-failure cycle of bearings.

RELIABILITY ENGINEERING & SYSTEM SAFETY (2022)

添加到收藏夹

Article Multidisciplinary Sciences

Uncertainty quantification of granular computing-neural network model for prediction of pollutant longitudinal dispersion coefficient in aquatic streams

Behzad Ghiasi et al.

Summary: Discharge of pollution loads into natural water systems remains a global challenge that threatens water and food supply, as well as endangering ecosystem services. This study develops an artificial intelligence-based predictive model that provides robust estimation of the rate of longitudinal dispersion (D-x) in streams under high spatiotemporal variability, addressing the large uncertainty in existing estimation models.

SCIENTIFIC REPORTS (2022)

添加到收藏夹

Article Biotechnology & Applied Microbiology

Audacity of huge: overcoming challenges of data scarcity and data quality for machine learning in computational materials discovery

Aditya Nandy et al.

Summary: Machine learning-accelerated discovery requires large amounts of high-fidelity data. However, the challenging nature and high cost of data generation in materials discovery have resulted in scarce and dubious quality data. Data-driven techniques such as using consensus across functionals, developing new theories, and detecting areas where computationally demanding methods are necessary are helping overcome these limitations. When properties cannot be reliably simulated, large experimental data sets can be used to train ML models. Moreover, advancements in natural language processing and automated image analysis enable the learning of structure-property relationships from literature.

CURRENT OPINION IN CHEMICAL ENGINEERING (2022)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Self-supervised learning methods and applications in medical imaging analysis: a survey

Saeed Shurrab et al.

Summary: This article reviews the application of self-supervised learning methods in medical imaging analysis, including predictive, generative, and contrastive approaches. It also covers recent research papers in the field and proposes possible future research directions.

PEERJ COMPUTER SCIENCE (2022)

添加到收藏夹

Article Automation & Control Systems

A robust model selection framework for fault detection and system health monitoring with limited failure examples: Heterogeneous data fusion and formal sensitivity bounds

Roberto Rocchetta et al.

Summary: This paper proposes a new model selection framework for fault detection and system health monitoring in engineering systems. The framework combines heterogeneous sources of information, a model of the system structure, and mathematical bounds on the sensitivity and specificity of fault classifiers. The results of three case studies demonstrate the efficacy and usefulness of the proposed approach.

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE (2022)

添加到收藏夹

Article Computer Science, Theory & Methods

A Survey of Deep Active Learning

Pengzhen Ren et al.

Summary: Researchers have shown relatively lower interest in active learning compared to deep learning, but with the increasing demand for large-scale high-quality annotated datasets, active learning is receiving more attention. This article provides a comprehensive survey on deep active learning, including a formal classification method, an overview of existing work, and an analysis of developments from an application perspective.

ACM COMPUTING SURVEYS (2022)

添加到收藏夹

Article Automation & Control Systems

Novel virtual sample generation using conditional GAN for developing soft sensor with small data

Qun-Xiong Zhu et al.

Summary: A new virtual sample generation approach CGAN-VSG is proposed to address the issue of insufficient data in data-driven soft sensing modeling of industrial processes. By integrating LOF with K-means++ algorithm and utilizing CGAN to generate virtual input and output samples, the proposed method enhances the accuracy of soft sensors with small data. Simulation results demonstrate the superiority of CGAN-VSG in terms of accuracy compared to other state-of-the-art methods.

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE (2021)

添加到收藏夹

Article Engineering, Industrial

Predictive maintenance enabled by machine learning: Use cases and challenges in the automotive industry

Andreas Theissler et al.

Summary: Recent advancements in maintenance modeling, particularly driven by data-based approaches like machine learning, have led to a wide range of applications in various industries. In the automotive sector, ensuring functional safety and optimizing maintenance costs has become a significant challenge. The lack of current surveys on ML-based predictive maintenance for automotive systems highlights the importance of understanding this emerging field. Key findings suggest that the availability of public data can boost research activities, combining multiple data sources can enhance accuracy, and the use of deep learning methods is expected to rise further.

RELIABILITY ENGINEERING & SYSTEM SAFETY (2021)

添加到收藏夹

Review Engineering, Multidisciplinary

A Survey on Contrastive Self-Supervised Learning

Ashish Jaiswal et al.

Summary: Self-supervised learning, particularly through contrastive learning, has gained popularity for its cost-effective approach in using self-defined pseudolabels for various downstream tasks. This paper extensively reviews self-supervised methods following the contrastive approach, explaining pretext tasks and different architectures used. Performance comparisons across multiple downstream tasks demonstrate variations in method effectiveness.

TECHNOLOGIES (2021)

添加到收藏夹

Article Economics

Recurrent Neural Networks for Time Series Forecasting: Current status and future directions

Hansika Hewamalage et al.

Summary: Although Recurrent Neural Networks (RNNs) have shown competitive forecasting performance, they are not always superior to established statistical models such as exponential smoothing (ETS) and autoregressive integrated moving average (ARIMA).

INTERNATIONAL JOURNAL OF FORECASTING (2021)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

Adaptive activation functions accelerate convergence in deep and physics-informed neural networks

Ameya D. Jagtap et al.

JOURNAL OF COMPUTATIONAL PHYSICS (2020)

添加到收藏夹

Article Automation & Control Systems

Potential, challenges and future directions for deep learning in prognostics and health management applications

Olga Fink et al.

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE (2020)

添加到收藏夹

Article Engineering, Industrial

Towards multi-model approaches to predictive maintenance: A systematic literature survey on diagnostics and prognostics

Juan Jose Montero Jimenez et al.

JOURNAL OF MANUFACTURING SYSTEMS (2020)

添加到收藏夹

Article Automation & Control Systems

Improving maritime traffic emission estimations on missing data with CRBMs

Alberto Gutierrez-Torre et al.

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE (2020)

添加到收藏夹

Article Engineering, Industrial

Remaining useful life predictions for turbofan engine degradation using semi-supervised deep architecture

Andre Listou Ellefsen et al.

RELIABILITY ENGINEERING & SYSTEM SAFETY (2019)

添加到收藏夹

Article Automation & Control Systems

Deep Transfer Learning Based on Sparse Autoencoder for Remaining Useful Life Prediction of Tool in Manufacturing

Chuang Sun et al.

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS (2019)

添加到收藏夹

Article Engineering, Environmental

Data scarcity in modelling and simulation of a large-scale WWTP: Stop sign or a challenge

Sina Borzooei et al.

JOURNAL OF WATER PROCESS ENGINEERING (2019)

添加到收藏夹

Article Engineering, Industrial

Bearing remaining useful life prediction based on deep autoencoder and deep neural networks

Lei Ren et al.

JOURNAL OF MANUFACTURING SYSTEMS (2018)

添加到收藏夹

Editorial Material Mathematical & Computational Biology

Recent Developments in Deep Learning for Engineering Applications

Athanasios Voulodimos et al.

COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE (2018)

添加到收藏夹

Article Mathematics, Interdisciplinary Applications

Predicting the Remaining Useful Life of an Aircraft Engine Using a Stacked Sparse Autoencoder with Multilayer Self-Learning

Jian Ma et al.

COMPLEXITY (2018)

添加到收藏夹

Article Computer Science, Information Systems

Residual Recurrent Neural Networks for Learning Sequential Representations

Boxuan Yue et al.

INFORMATION (2018)

添加到收藏夹

Review Engineering, Multidisciplinary

Prognostics and Health Management: A Review on Data Driven Approaches

Kwok L. Tsui et al.

MATHEMATICAL PROBLEMS IN ENGINEERING (2015)

添加到收藏夹

Review Multidisciplinary Sciences

Deep learning

Yann LeCun et al.

NATURE (2015)

添加到收藏夹

Article Automation & Control Systems

Prediction intervals to account for uncertainties in neural network predictions: Methodology and application in bus travel time prediction

Ehsan Mazloumi et al.

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE (2011)

添加到收藏夹

Article Computer Science, Hardware & Architecture

Exploiting mobility for energy efficient data collection in wireless sensor networks

S Jain et al.

MOBILE NETWORKS & APPLICATIONS (2006)

添加到收藏夹

© Peeref 2019-2024. All rights reserved.