Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
3
result(s) for
"Machine learining."
Sort by:
Smart proxy modeling : artificial intelligence and machine learning in numerical simulation
\"Numerical simulation models are used in all engineering disciplines for modeling physical phenomena to learn how the phenomena work, and to identify problems and optimize behavior. Smart proxy models provide an opportunity to replicate numerical simulations with very high accuracy and can be run on a laptop within a few minutes, thereby simplifying the use of complex numerical simulations which can otherwise take tens of hours. This book focuses on smart proxy modeling and provides readers with all the essential details on how to develop smart proxy models using artificial intelligence and machine learning, as well as how it may be used in real-world cases. Covers replication of highly accurate numerical simulations using artificial intelligence and machine learning. Details application in reservoir simulation and modeling, and computational fluid dynamics. Includes real case studies based on commercially available simulators. Smart Proxy Modeling is ideal for petroleum, chemical, environmental, and mechanical engineers, as well as statisticians and others working with applications of data-driven analytics\"-- Provided by publisher.
Book
Tonsil Mycobiome in PFAPA (Periodic Fever, Aphthous Stomatitis, Pharyngitis, Adenitis) Syndrome: A Case-Control Study
Periodic fever, aphthous stomatitis, pharyngitis and adenitis syndrome (PFAPA) is the most common periodic fever syndrome in children with unknown etiology, effectively treated with tonsillectomy. Earlier we have shown that tonsil microbiome is different in patients with PFAPA as compared to that in controls. Recently, fungal microbiome, mycobiome, has been linked to the pathogenesis of inflammatory diseases. We now investigated the role of mycobiome of tonsils in PFAPA. Random forest classification, a machine learning approach, was used for the analysis of mycobiome data. We examined tonsils from 30 children with PFAPA and 22 control children undergoing tonsillectomy for non-infectious reasons. We identified 103 amplicon sequence variants, mainly from two fungal phyla, Ascomycota and Basidiomycota. The mean relative abundance of Candida albicans in the tonsil mycobiome was 11% (95% CI: 19 to 27%) in cases and 3.4 % (95% CI: -0.8% to 8%) in controls, p =0.104. Mycobiome data showed no statistical difference in differentiating between PFAPA cases and controls compared to a random chance classifier (area under the curve (AUC) = 0.47, SD = 0.05, p = 0.809). In conclusion, in this controlled study, tonsillar mycobiome in children with PFAPA syndrome did not differ from that of the controls.
Journal Article
A viable framework for semi-supervised learning on realistic dataset
Semi-supervised Fine-Grained Recognition is a challenging task due to the difficulty of data imbalance, high inter-class similarity and domain mismatch. Recently, this field has witnessed giant leap and many methods have gained great performance. We discover that these existing Semi-supervised Learning (SSL) methods achieve satisfactory performance owe to the exploration of unlabeled data. However, on the realistic large-scale datasets, due to the abovementioned challenges, the improvement of the quality of pseudo-labels requires further research. In this work, we propose Bilateral-Branch Self-Training Framework (BiSTF), a simple yet effective framework to improve existing semi-supervised learning methods on class-imbalanced and domain-shifted fine-grained data. By adjusting stochastic epoch update frequency, BiSTF iteratively retrains a baseline SSL model with a labeled set expanded by selectively adding pseudo-labeled samples from an unlabeled set, where the distribution of pseudo-labeled samples is the same as the labeled data. We show that BiSTF outperforms the existing state-of-the-art SSL algorithm on Semi-iNat dataset. Our code is available at
https://github.com/HowieChangchn/BiSTF
.
Journal Article