Particulate discrimination via multimodal machine learning in coordinated space
Explore the source record for details and available documents.
Deneke, Elihu (ORCID:0000000190912916)↗
Engineering topics
Miller, Jonathan
Publications and source records attributed to Miller, Jonathan.
Inferring Convolutional Neural Networks' Accuracies from Their Architectural Characterizations
The challenge of choosing an appropriate convolutional neural network (CNN) architecture for specific applications and different data sets is still poorly understood in the literature. This is problematic, since CNNs have shown strong promise for analyzing scientific data from many domains including particle imaging detectors. In this paper, we proposed a systematic language that is useful for comparison between different CNN's architectures before training time. This helps us predict whether a network can perform better than a certain threshold accuracy before training up to 70% accuracy using simple machine learning models. Additionally, we found a coefficient of determination of 0.966 for an Ordinary Least Squares model in a regression task to predict accuracy of a large population of networks.
Hoang, Duc↗
The History & Evolution of Spacesuits and Spacewalks - Walking in Space
The presentation and videos that will be included in this technology talk will summarize the basic functions of spacesuits, the evolution of spacesuit design, and the development plans for future exploration spacesuits. The videos will run in a loop with no audio. The speakers will generally follow the slide presentation. There will be a 5-minute intro on basic suit functions, followed by a 7-10 minute discussion on suit history and evolution, then 7-10 minutes to cover the current ISS (International Space Station) suit and the development of the next generation exploration spacesuits. That will leave around 5-10 minutes for questions and answers.
Hughes, David↗
Evolving Deep Networks Using HPC
While a large number of deep learning networks have been studiedand published that produce outstanding results on natural imagedatasets, these datasets only make up a fraction of those to whichdeep learning can be applied. These datasets include text data, audiodata, and arrays of sensors that have very different characteristicsthan natural images. As these “best” networks for natural imageshave been largely discovered through experimentation and cannotbe proven optimal on some theoretical basis, there is no reasonto believe that they are the optimal network for these drasticallydifferent datasets. Hyperparameter search is thus often a very im-portant process when applying deep learning to a new problem. Inthis work we present an evolutionary approach to searching thepossible space of network hyperparameters and construction thatcan scale to 18, 000 nodes. This approach is applied to datasets ofvarying types and characteristics where we demonstrate the abilityto rapidly find best hyperparameters in order to enable practitionersto quickly iterate between idea and result.
Young, Steven↗
Thermal Infrared Radiometric Calibration of the Entire Landsat 4, 5, and 7 Archive (1982-2010)
Landsat's continuing record of the thermal state of the earth's surface represents the only long term (1982 to the present) global record with spatial scales appropriate for human scale studies (i.e., tens of meters). Temperature drives many of the physical and biological processes that impact the global and local environment. As our knowledge of, and interest in, the role of temperature on these processes have grown, the value of Landsat data to monitor trends and process has also grown. The value of the Landsat thermal data archive will continue to grow as we develop more effective ways to study the long term processes and trends affecting the planet. However, in order to take proper advantage of the thermal data, we need to be able to convert the data to surface temperatures. A critical step in this process is to have the entire archive completely and consistently calibrated into absolute radiance so that it can be atmospherically compensated to surface leaving radiance and then to surface radiometric temperature. This paper addresses the methods and procedures that have been used to perform the radiometric calibration of the earliest sizable thermal data set in the archive (Landsat 4 data). The completion of this effort along with the updated calibration of the earlier (1985 1999) Landsat 5 data, also reported here, concludes a comprehensive calibration of the Landsat thermal archive of data from 1982 to the present
Schott, John R.↗