A5026172631- Planetary Science and Exploration
- Astro and Planetary Science
- Space Exploration and Technology
- Anomaly Detection Techniques and Applications
- Radio Astronomy Observations and Technology
- Data Management and Algorithms
- Machine Learning and Data Classification
- Advanced Clustering Algorithms Research
- Geochemistry and Geologic Mapping
- Scientific Computing and Data Management
- Image Processing and 3D Reconstruction
- Spacecraft Design and Technology
- Space Science and Extraterrestrial Life
- Remote-Sensing Image Classification
- Pulsars and Gravitational Waves Research
- Algorithms and Data Compression
- Astrophysics and Cosmic Phenomena
- Data Mining Algorithms and Applications
- Machine Learning and Algorithms
- Space Satellite Systems and Control
- Image Retrieval and Classification Techniques
- Explainable Artificial Intelligence (XAI)
- Isotope Analysis in Ecology
- Data Stream Mining Techniques
- Bayesian Methods and Mixture Models
Independent Sector
2025
Jet Propulsion Laboratory
2015-2024
Oregon State University
2024
California Institute of Technology
2013-2022
San Jose State University
2017-2018
Pasadena City College
2017
University of Southern California
2010
Tufts University
2009
University of California, Davis
2008
Johns Hopkins University Applied Physics Laboratory
2005
Abstract The Australian Square Kilometre Array Pathfinder (ASKAP) will give us an unprecedented opportunity to investigate the transient sky at radio wavelengths. In this paper we present VAST, ASKAP survey for Variables and Slow Transients. VAST exploit wide-field capabilities of enable discovery investigation variable phenomena from local cosmological, including flare stars, intermittent pulsars, X-ray binaries, magnetars, extreme scattering events, interstellar scintillation, supernovae,...
Abstract The InSight lander represents a unique opportunity to correlate seismic data with impact events identified in orbital images, enabling the characterization of physical properties martian crust and mantle. Here, we present first comprehensive catalog impacts that occurred during mission within 50° radius around lander. We use machine learning‐enabled approach identify 123 date‐constrained diameters between ∼1 22.5 m. estimate an rate 2.7 × 10 −6 /km 2 /year for >3.9 m effective...
Abstract We are developing a purely commensal survey experiment for fast (<5 s) transient radio sources. Short-timescale transients associated with the most energetic and brightest single events in Universe. Our objective is to cover enormous volume of parameter space made available by ASKAP, an unprecedented combination sensitivity field view. Fast timescale open new vistas on physics high brightness temperature emission, extreme states matter strong gravitational fields. In addition,...
Abstract We present a catalog of new impacts on Mars. These craters formed in the last few decades, constrained with repeat orbital imaging. Crater diameters range from 58 m down to <1 m. For each impact, we report whether it single crater or cluster (58% clusters); albedo features blast zone (88% halos; 64% linear rays; 10% arcuate majority dark‐toned; 4% light‐toned; 14% dual‐toned); and exposures ice (4% definite; 2% possible). find no trends occurrences clusters latitude, elevation,...
Despite significant technical advances in machine learning (ML) over the past several years, tangible impact of this technology healthcare has been limited. This is due not only to particular complexities healthcare, but also structural issues for (MLHC) community which broadly reward novelty tangible, equitable impact. We structure our work as a healthcare-focused echo 2012 paper “Machine Learning that Matters”, highlighted such ML at large, and offered series clearly defined “Impact...
This article is a part of the Special Issue on Intelligent Systems for Space Exploration. The Payload Experiment (IPEX) CubeSat that flew from December 2013 through January 2015 and validated autonomous operations onboard instrument processing product generation Module Hyperspectral Infrared Imager (HyspIRI) mission concept. IPEX used several artificial intelligence technologies. First, machine learning computer vision in its processing. machine-learned random decision forests to classify...
We present and evaluate the initial version of RIPTIDES, a system that combines information extraction, extraction-based summarization, natural language generation to support user-directed multidocument summarization.
Recent discoveries of dispersed, non-periodic impulsive radio signals with single-dish telescopes have sparked significant interest in exploring the relatively uncharted space fast transient signals. Here we describe V-FASTR, an experiment to perform a blind search for using Very Long Baseline Array (VLBA). The runs entirely commensal mode, alongside normal VLBA observations and operations. It is made possible by features flexibility DiFX software correlator that used process data. Using...
NASA has acquired more than 22 million images from the planet Mars. To help users find of interest, we developed a content-based search capability for Mars rover surface and orbital images. We started with AlexNet convolutional neural network, which was trained on Earth images, used transfer learning to adapt network use report our deployment these classifiers within PDS Imaging Atlas, publicly accessible web interface, enable first image NASA’s
In this work, we present a system based on convolutional autoencoders for detecting novel features in multispectral images. We introduce SAMMIE: Selections Autoencoder Modeling of Multispectral Image Expectations. Previous work using employed the scalar reconstruction error to classify new images as or typical. show that spatial-spectral map can enable both accurate classification novelty well human-comprehensible explanations detection. apply our methodology detection geologic Martian...
Ongoing planetary exploration missions are returning large volumes of image data. Identifying surface changes in these images, e.g., new impact craters, is critical for investigating many scientific hypotheses. Traditional approaches to change detection rely on differencing and manual feature engineering. These methods can be sensitive irrelevant variations illumination or quality typically require before after images coregistered, which itself a major challenge. Additionally, most prior...
Time domain radio astronomy observing campaigns frequently generate large volumes of data. Our goal is to develop automated methods that can identify events interest buried within the larger data stream. The V-FASTR fast transient system was designed detect rare bursts collected by Very Long Baseline Array. resulting event candidates constitute a significant burden in terms subsequent human reviewing time. We have trained and deployed machine learning classifier marks each candidate...
Abstract Science teams for rover-based planetary exploration missions like the Mars Laboratory Curiosity rover have limited time analyzing new data before making decisions about follow-up observations. There is a need systems that can rapidly and intelligently extract information from instrument datasets focus attention on most promising or novel Several novelty detection methods been explored in prior work three-channel color images non-image datasets, but few considered multispectral...
Article Alpha seeding for support vector machines Share on Authors: Dennis DeCoste Machine Learning Systems Group, Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Drive, Pasadena, CA CAView Profile , Kiri Wagstaff Department Computer Science, Cornell University, 4156 Upson Hall, Ithaca, NY NYView Authors Info & Claims KDD '00: Proceedings the sixth ACM SIGKDD international conference Knowledge discovery and data miningAugust 2000 Pages...