20 Aug 2020

PUBLICATION NOTICE: Site-Specific Case Studies for Determining Ground Snow Loads in the United States

ABSTRACT:  The U.S. Army Cold Regions Research and Engineering Laboratory (CRREL) has mapped ground snow loads for much of the United States. In some areas where extreme local variations preclude mapping on a national scale, instead of loads, “CS” is used to indicate that Case Studies are needed. This report and the accompanying spreadsheet, which contains the 15,104-station CRREL ground snow load database, provide the information needed to conduct Case Studies. When the latitude, longitude, and elevation of a site of interest are provided, the spreadsheet tabulates data available in the vicinity and generates plots that relate ground snow loads nearby to elevation. With this information, the ground snow load at the site of interest can be determined. This report uses 10 examples to illustrate the methodology and provides our answer and the comments we generate for each of these Case Studies and for 16 additional sites of interest, 8 of which have their answers “disguised” for practice purposes. CRREL has conducted over 1000 Case Studies upon request. Practicing structural engineers were involved in over 250 of them to verify that this methodology is ready to transfer to the design profession.

20 Aug 2020

PUBLICATION NOTICE: Understanding State-of-the-Art Material Classification through Deep Visualization

Abstract: Neural networks (NNs) excel at solving several complex, non-linear problems in the area of supervised learning. A prominent application of these networks is image classification. Numerous improvements over the last few decades have improved the capability of these image classifiers. However, neural networks are still a black-box for solving image classification and other sophisticated tasks. A number of experiments conducted look into exactly how neural networks solve these complex problems. This paper dismantles the neural network solution, incorporating convolution layers, of a specific material classifier. Several techniques are utilized to investigate the solution to this problem. These techniques look at specifically which pixels contribute to the decision made by the NN as well as a look at each neuron’s contribution to the decision. The purpose of this investigation is to understand the decision-making process of the NN and to use this knowledge to suggest improvements to the material classification algorithm.

20 Aug 2020

PUBLICATION NOTICE: Understanding State-of-the-Art Material Classification through Deep Visualization

Abstract: Neural networks (NNs) excel at solving several complex, non-linear problems in the area of supervised learning. A prominent application of these networks is image classification. Numerous improvements over the last few decades have improved the capability of these image classifiers. However, neural networks are still a black-box for solving image classification and other sophisticated tasks. A number of experiments conducted look into exactly how neural networks solve these complex problems. This paper dismantles the neural network solution, incorporating convolution layers, of a specific material classifier. Several techniques are utilized to investigate the solution to this problem. These techniques look at specifically which pixels contribute to the decision made by the NN as well as a look at each neuron’s contribution to the decision. The purpose of this investigation is to understand the decision-making process of the NN and to use this knowledge to suggest improvements to the material classification algorithm.

19 Aug 2020

MKARNS Nav Notice SWL 20-50 Lock 3 Sailing Instruction Lifted

MKARNS - The sailing instruction for the downstream approach to Joe Hardin Lock (No. 3) NM 50.2, as noted in Navigation Notice SWL 20-42 has been lifted.