Search Results

You are looking at 41 - 50 of 857 items for :

  • Deep learning x
  • Refine by Access: All Content x
Clear All
Stephan Rasp and Sebastian Lerch

parameters without having to specify appropriate link functions, and the ease of adding station information into a global model by using embeddings. The network model parameters are estimated by optimizing the CRPS, a nonstandard choice in the machine learning literature tailored to probabilistic forecasting. Furthermore, the rapid pace of development in the deep learning community provides flexible and efficient modeling techniques and software libraries. The presented approach can therefore be easily

Open access
Buo-Fu Chen, Boyo Chen, Hsuan-Tien Lin, and Russell L. Elsberry

various life stages, environments, and basins. In addition, only a few features (usually less than 10) may be finally used in the regression models. This collaborative study between meteorologists and data scientists proposes a deep-learning model to address the need for an automated, objective, and end-to-end intensity estimation technique. Since AlexNet, which established the baseline architecture of convolutional neural networks for image recognition used today, was proposed in 2012 ( Krizhevsky

Full access
Kirkwood A. Cloud, Brian J. Reich, Christopher M. Rozoff, Stefano Alessandrini, William E. Lewis, and Luca Delle Monache

-scale meteorological predictors, along with information describing the atmospheric flow stability and the uncertainty in initial conditions, to predict forecast intensity error in operational prediction schemes. A18 also addressed intensity prediction with the analog ensemble method. More recently, machine learning has gained increasing prominence in postprocessing. Evolutionary programming, simple neural networks, and deep learning have shown significant promise as postprocessing tools (e.g., Gagne et al. 2014

Full access
Jonathan D. W. Kahl

The Atmospheric Science program at the University of Wisconsin—Milwaukee regularly offers the general education course Survey of Meteorology, serving over 400 students each year. This article describes a systematic inquiry into the teaching and learning goals of the course and the adequacy of current methods used to assess student performance. Following a survey of the six faculty members with teaching responsibilities for the course, common student learning goals of meteorological content and the application of meteorological concepts to observations were identified. Student surveys, designed to assess both the extent to which these learning goals were being met as well as the depth of learning, were administered to 241 students during the 2005–06 academic year. Results indicate that 80% of students surveyed met the content learning goal, while the application learning goal was met by only 66% of students. A deeper level of application learning, involving pattern recognition and the separation of concepts into component parts, was achieved by only 45% of the students. A comparison of student survey results with course grade distributions indicates that current grading practices are adequate for assessing the content learning goal but are inadequate for assessing the application learning goal.

Full access
Melanie M. Colavito, Sarah F. Trainor, Nathan P. Kettle, and Alison York

time to facilitate the development and application of actionable science is needed ( McNie 2013 ; Meadow et al. 2015 ; Wall et al. 2017b ). Table 1. Key modes of science production in boundary organizations as used in this paper. This paper uses case study research to assess the Alaska Fire Science Consortium (AFSC)—a boundary organization focused on fire science and management in Alaska—and to develop a deeper understanding of how boundary organizations and knowledge coproduction work in

Full access
Peter J. Walton, Morgan B. Yarker, Michel D. S. Mesquita, and Friederike E. L. Otto

; however, having enough bandwidth to be able to download can be problematic to some participants. This course structure though does reflect more closely Kolb’s (1984) model of experiential learning, where it is suggested that participants will form deeper understanding concepts (in this case climate modeling) through actively participating in an exercise. The final point to be taken from comparing the two courses is the issue of certification. As well as wanting to develop their own knowledge

Full access
David R. Perkins IV, Jennifer Vanos, Christopher Fuhrmann, Michael Allen, David Knight, Cameron C. Lee, Angela Lees, Andrew Leung, Rebekah Lucas, Hamed Mehdipoor, Sheila Tavares Nascimento, Scott Sheridan, and Jeremy Spencer

learning, a natural research inquiry surfaced: how do we enhance the teaching and learning of biometeorology in higher education to both promote the discipline and elicit deeper learning into common meteorological, climatological, and biological concepts? As a result of this inquiry, the second international workshop held by the ISB’s SNP group sought to tackle this very issue. Transforming core tenants of biometeorology and education into collaborative initiatives on an international scale was a key

Open access
Lisa M. PytlikZillig, Qi Hu, Kenneth G. Hubbard, Gary D. Lynne, and Roger H. Bruning

explanations are essential for enhancing their ownership of the new knowledge they are developing. In addition, explanation also can facilitate the learning of new skills, deeper learning, and better integration of new knowledge with prior knowledge ( Ainsworth and Loizou 2003 ; Chi et al. 1989 , 1994 ; Renkl 2002 ; Roscoe and Chi 2008 ; Roy and Chi 2005 ). Meanwhile, peer discussion can enhance many important outcomes, including understanding, critical thinking, and construction of complex knowledge

Full access
A. Hicks and B. M. Notaroš

they are capable of transferring learning from one dataset to another and are not limited to specific parameters inherent either to the dataset (e.g., resolution, color, or size) or capturing method (e.g., hardware imperfections reflected in data). Therefore, a classifier properly trained with a CNN can be utilized by a variety of image-capturing in situ devices. Research into deep learning has extended their ability to process complex data without major changes to the algorithm. Finally, CNNs are

Full access
Sid-Ahmed Boukabara, Vladimir Krasnopolsky, Stephen G. Penny, Jebb Q. Stewart, Amy McGovern, David Hall, John E. Ten Hoeve, Jason Hickey, Hung-Lung Allen Huang, John K. Williams, Kayo Ide, Philippe Tissot, Sue Ellen Haupt, Kenneth S. Casey, Nikunj Oza, Alan J. Geer, Eric S. Maddy, and Ross N. Hoffman

observations and satellite applications, including the use of neural networks for NWP model parameterization ( Krasnopolsky et al. 2010 ) and using deep learning to infer missing data ( Boukabara et al. 2019a ). NCAR, a federally funded research and development center, has a long history of developing AI techniques for weather applications. Haupt et al. (2019) highlighted the Dynamic Integrated Forecast (DICast) system, a 20-year effort at NCAR that forms the “weather engine” of many applications, as

Open access