Particulate discrimination via multimodal machine learning in coordinated space
Explore the source record for details and available documents.
Engineering topics
Publications and source records attributed to Miller, Jonathan.
Explore the source record for details and available documents.
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.
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.
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.
MINERvA (Main INjector ExpeRiment for v-A) is a few-GeV neutrino nucleus scattering experiment at Fermilab using various nuclei as targets. The experiment provides measurements of neutrino and anti-neutrino cross sections off of nuclear targets which are important for neutrino oscillation experiments and the probing of the nuclear medium.Presented are recent results from MINERvA on quasi-elastic, inclusive charged-current neutrino scattering, and pion production processes.
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
The electric form factor of the neutron was determined from studies of the reaction \rea{} in quasi-elastic kinematics in Hall A at Jefferson Lab. Longitudinally polarized electrons were scattered off a polarized target in which the nuclear polarization was oriented perpendicular to the momentum transfer. The scattered electrons were detected in a magnetic spectrometer in coincidence with neutrons that were registered in a large-solid-angle detector. More than doubling the $Q^2$-range over which it is known, we find \GEn{}$ = 0.0225 \pm 0.0017 (stat) \pm 0.0024 (syst)$, $0.0200 \pm 0.0023 \pm 0.0018$, and $0.0142 \pm 0.0019 \pm 0.0013$ for $Q^2$ = 1.72, 2.48, and 3.41~\gevsq, respectively.
The E02-013 collaboration's precision measurement of the electric form factor of the neutron at high momentum took place in Hall A of Jefferson Laboratory [1]. Four kinematic points spread in Q2 from 1.2 to 3.5 (GeV/c)2 reach the region of momentum transfer where there is little ambiguity that the form factors are dominated by the valence quarks. The electric form factor provides important constraints on the generalized parton distributions (GPDs) of the nucleon and therefore to understand the currently unknown quark angular orbital moments. The talk presented newly obtained results for GnE including at Q2 of 2.5 (GeV/c)2. These measurements provide access to twice the range of momentum transfer for complete iso-spin decomposition of the nucleon form factors. The form factors of the u and d quarks were also presented. The measurement used a double polarization asymmetry method [2][3] with He polarized up to 55% (provided by a novel hybrid alkali optical pumping scheme utilizing a Rb & K mixture) and the highly polarized (85%) electron beam at CEBAF. A specially constructed detector package consisting of an electron spectrometer, named Big-Bite, with a solid angle of 95 msr and a large neutron detector were used to detect the particles in the reaction {sup 3}H{rvec e}({rvec e},e',h). The experimental apparatus provided more than 100 times better Figure-of-Merit than other GnE experiments utilizing polarized targets; for more information see [4].
The E02-013 collaboration s precision measurement of the electric form factor of the neutron at high momentum took place in Hall A of Jefferson Laboratory [1]. Four kinematic points spread in Q2 from 1.2 to 3.5 (GeV/c)2 reach the region of momentum transfer where there is little ambiguity that the form factors are dominated by the valence quarks. The electric form factor provides important constraints on the generalized parton distributions (GPDs) of the nucleon and therefore to understand the currently unknown quark angular orbital moments. The talk presented newly obtained results for GnE including at Q2 of 2.5 (GeV/c)2. These measurements provide access to twice the range of momentum transfer for complete iso-spin decomposition of the nucleon form factors. The form factors of the u and d quarks were also presented. The measurement used a double polarization asymmetry method [2][3] with He polarized up to 55% (provided by a novel hybrid alkali optical pumping scheme utilizi