General-Purpose Bayesian Tensor Learning With Automatic Rank Determination and Uncertainty Quantification

Zhang, Kaiqi; Hawkins, Cole; Zhang, Zheng

doi:10.3389/frai.2021.668353

Cited by 6 publications

(4 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The curse of dimensionality has always been an issue in the ML research ( Quintero et al, 2021 ; Zhang et al, 2021 ). For classification, there are not enough data objects to create a model to reliably assign all possible objects to a class.…”

Section: Resultsmentioning

confidence: 99%

Machine learning algorithm performance evaluation in structural magnetic resonance imaging-based classification of pediatric bipolar disorders type I patients

Dou

Gao

Meng

et al. 2022

Front. Comput. Neurosci.

View full text Add to dashboard Cite

The diagnosis based on clinical assessment of pediatric bipolar disorder (PBD) may sometimes lead to misdiagnosis in clinical practice. For the past several years, machine learning (ML) methods were introduced for the classification of bipolar disorder (BD), which were helpful in the diagnosis of BD. In this study, brain cortical thickness and subcortical volume of 33 PBD-I patients and 19 age-sex matched healthy controls (HCs) were extracted from the magnetic resonance imaging (MRI) data and set as features for classification. The dimensionality reduced feature subset, which was filtered by Lasso or f_classif, was sent to the six classifiers (logistic regression (LR), support vector machine (SVM), random forest classifier, naïve Bayes, k-nearest neighbor, and AdaBoost algorithm), and the classifiers were trained and tested. Among all the classifiers, the top two classifiers with the highest accuracy were LR (84.19%) and SVM (82.80%). Feature selection was performed in the six algorithms to obtain the most important variables including the right middle temporal gyrus and bilateral pallidum, which is consistent with structural and functional anomalous changes in these brain regions in PBD patients. These findings take the computer-aided diagnosis of BD a step forward.

show abstract

Section: Resultsmentioning

confidence: 99%

Machine learning algorithm performance evaluation in structural magnetic resonance imaging-based classification of pediatric bipolar disorders type I patients

Dou

Gao

Meng

et al. 2022

Front. Comput. Neurosci.

View full text Add to dashboard Cite

show abstract

“…Graph Attention Network (GAT) [15] is a spatial-based GNN model, specifying respective weights to each node within a neighborhood via Attention Mechanism, comparing with GraphSAGE and many other popular GNNs regarded all neighbors of a node as equal importance [14]. Moreover, Gated Attention Networks (GaAN) [30] computes the attention score of each attention head instead of assuming each attention head has an equal contribution. Furthermore, GNNs has also been applied to conduct various classification tasks.…”

Section: 2graph Neural Network (Gnns)mentioning

confidence: 99%

Prediction of heart disease using graph neural networks

Chen

2023

International Conference on Artificial Intelligence and Industrial Design (AIID 2022)

View full text Add to dashboard Cite

Cardiovascular Diseases (CVDs) have become increasingly crucial in recent years and have been regarded as the leading cause of death worldwide. Although it is necessary to detect and treat CVDs in their early stages, only 67% of heart diseases could be predicted by medical professionals. Motivated by recent advances in Graph Neural Networks (GNNs) that have ramified in a variety of industries, in this paper, we utilize three novel GNN models, including Graph Convolutional Network (GCN), GraphSAGE, and Graph Attention Network (GAT), to conduct the heart disease prediction task, comprised of two stages: table-to-graph transformation and Graph Neural Networks prediction. Experimental results show significant improvements with the utilization of GNNs compared with three novel Machine Learning models: Logistic Regression (LR), Naïve Bayes (NB), and Multi-Layer Perceptron (MLP), and GAT performs optimally among the three GNN models. To the best of our knowledge, this is the first work that predicts heart disease using GNNs.

show abstract

“…Following the rise of the real-world applications of machine learning, the concept of uncertainty quanti cation emerged as an important component to better analyze and understand predictive models [72][73][74]. Quantifying the uncertainties in a machine learning model depends on the underlying forecasting task, as well as the settings of the model itself.…”

Section: Uncertainty Quanti Cationmentioning

confidence: 99%

Downscaling daily wind speed with Bayesian deep learning for climate monitoring

Gerges

Boufadel

Bou‐Zeid

et al. 2023

Int J Data Sci Anal

View full text Add to dashboard Cite

Wind dynamics are extremely complex and have critical impacts on the level of damage from natural hazards, such as storms and wild res. In the wake of climate change, wind dynamics are becoming more complex, making the prediction of future wind characteristics a more challenging task. Nevertheless, having long-term projections of some wind characteristics, such as daily wind speed, is crucial for effective monitoring of climate change, and for e cient disaster risk management. Furthermore, accurate projections of wind speed result in optimized generation of wind-based electric power. General Circulation Models (GCMs) provide long-term simulations (often till year 2100 or more) of multiple climate variables. However, simulations from a GCM are at a grid with coarse spatial resolution, rendering them ineffective to resolve and analyze climate change at the local regional level. Spatial downscaling techniques are often used to map such global large-scale simulations to a local small-scale region. In this paper, we present a novel deep learning framework for spatial downscaling, speci cally for forecasting the daily average wind speed at a local station level using GCM simulations. Our framework, named Wind Convolutional Neural Networks with Transformers, or WCT for short, consists of multi-head convolutional neural networks, followed by stacked transformers, and an uncertainty quanti cation component based on Bayesian inference. Experimental results show the suitability of WCT when applied on four wind stations in New Jersey and Pennsylvania, United States. Moreover, we use the trained WCT on future GCM simulations to produce local-scale daily wind speed projections up to year 2100.

show abstract

General-Purpose Bayesian Tensor Learning With Automatic Rank Determination and Uncertainty Quantification

Cited by 6 publications

References 22 publications

Machine learning algorithm performance evaluation in structural magnetic resonance imaging-based classification of pediatric bipolar disorders type I patients

Machine learning algorithm performance evaluation in structural magnetic resonance imaging-based classification of pediatric bipolar disorders type I patients

Prediction of heart disease using graph neural networks

Downscaling daily wind speed with Bayesian deep learning for climate monitoring

Contact Info

Product

Resources

About