A5076599178- Superconducting Materials and Applications
- Advancements in Semiconductor Devices and Circuit Design
- Particle accelerators and beam dynamics
- Particle Accelerators and Free-Electron Lasers
- Gamma-ray bursts and supernovae
- Electrical Fault Detection and Protection
- Advanced Optical Sensing Technologies
- Astronomical Observations and Instrumentation
- Radiation Detection and Scintillator Technologies
- Non-Destructive Testing Techniques
- Time Series Analysis and Forecasting
- Anomaly Detection Techniques and Applications
- Astrophysics and Cosmic Phenomena
- Astrophysical Phenomena and Observations
- Astronomy and Astrophysical Research
- Stellar, planetary, and galactic studies
- Spectroscopy and Laser Applications
- Image Processing Techniques and Applications
- Fault Detection and Control Systems
- Mineral Processing and Grinding
- Galaxies: Formation, Evolution, Phenomena
- Remote-Sensing Image Classification
- Advanced Chemical Sensor Technologies
- Advanced Fluorescence Microscopy Techniques
- Scientific Computing and Data Management
Millennium Institute of Astrophysics
2017-2024
Data Observatory Foundation
2024
University of Chile
2016-2024
University of Bío-Bío
2023
Fermi National Accelerator Laboratory
2010
University of Chicago
2010
Abstract We introduce Deep-HiTS, a rotation-invariant convolutional neural network (CNN) model for classifying images of transient candidates into artifacts or real sources the High cadence Transient Survey (HiTS). CNNs have advantage learning features automatically from data while achieving high performance. compare our CNN against feature engineering approach using random forests (RFs). show that significantly outperforms RF model, reducing error by almost half. Furthermore, fixed number...
We propose a new sequential classification model for astronomical objects based on recurrent convolutional neural network (RCNN) which uses sequences of images as inputs. This approach avoids the computation light curves or difference images. is first time that are used directly variable in astronomy. The second contribution this work image simulation process. generate synthetic take into account instrumental and observing conditions, obtaining realistic, set movies each object. simulated...
Abstract We present a real-time stamp classifier of astronomical events for the Automatic Learning Rapid Classification Events broker, ALeRCE. The is based on convolutional neural network, trained alerts ingested from Zwicky Transient Facility (ZTF). Using only science, reference, and difference images first detection as inputs, along with metadata alert features, able to correctly classify active galactic nuclei, supernovae (SNe), variable stars, asteroids, bogus classes, high accuracy...
ALeRCE (Automatic Learning for the Rapid Classification of Events) is currently processing Zwicky Transient Facility (ZTF) alert stream, in preparation Vera C. Rubin Observatory, and classifying objects using a broad taxonomy. The light curve classifier balanced random forest (BRF) algorithm with two-level scheme that uses variability features computed from ZTF colors obtained AllWISE photometry. This work develops an updated version broker includes tidal disruption events (TDEs) as new...
A significant challenge in the study of transient astrophysical phenomena is identification bogus events, among which human-made Earth-orbiting satellites and debris remain major contaminants. Existing pipelines can effectively identify satellite trails, but they often miss more complex signatures, such as collections glints. In Rubin Observatory era, scale operations will increase tenfold with respect to its precursor, Zwicky Transient Facility (ZTF), requiring crucial improvements...
During the last couple of years Astronomy, as many other fields, has been facing problem automatic processing massive data. This primarily driven by large survey instruments, which scan areas sky aiming to catalog everything they find. The High Cadence Transient Survey (HiTS) is one these surveys that searches for phenomena limited duration (such supernovae) called transients. HiTS pipeline produces transient candidates subtracting science images from a template. In this paper we present...
Abstract We present DELIGHT, or Deep Learning Identification of Galaxy Hosts Transients, a new algorithm designed to automatically and in real time identify the host galaxies extragalactic transients. The proposed receives as input compact, multiresolution images centered at position transient candidate outputs two-dimensional offset vectors that connect with center its predicted host. consists set same number pixels, but progressively larger pixel sizes fields view. A sample 16,791 visually...
We introduce the Automatic Learning for Rapid Classification of Events (ALeRCE) broker, an astronomical alert broker designed to provide a rapid and self--consistent classification large etendue telescope streams, such as that provided by Zwicky Transient Facility (ZTF) and, in future, Vera C. Rubin Observatory Legacy Survey Space Time (LSST). ALeRCE is Chilean--led run interdisciplinary team astronomers engineers, working become intermediaries between survey follow--up facilities. uses...
Context . The advent of next-generation survey instruments, such as the Vera C. Rubin Observatory and its Legacy Survey Space Time (LSST), is opening a window for new research in time-domain astronomy. Extended LSST Astronomical Time-Series Classification Challenge (ELAsTiCC) was created to test capacity brokers deal with simulated stream. Aims Our aim develop model classification variable astronomical objects. We describe ATAT, Transformer time series And Tabular data, conceived by ALeRCE...
In this paper, we propose an enhanced CNN model for detecting supernovae (SNe). This is done by applying a new method obtaining rotational invariance that exploits cyclic symmetry. addition, use visualization approach, the layerwise relevance propagation (LRP) method, which allows finding relevant pixels in each image contribute to discriminate between SN candidates and artifacts. We introduce measure assess quantitatively effect of invariant methods on LRP heatmaps. comparing proposed CAP,...
Abstract We present the first version of Automatic Learning for Rapid Classification Events (ALeRCE) broker light curve classifier. ALeRCE is currently processing Zwicky Transient Facility (ZTF) alert stream, in preparation Vera C. Rubin Observatory. The classifier uses variability features computed from ZTF stream and colors obtained AllWISE photometry. apply a balanced random forest algorithm with two-level scheme where top level classifies each source as periodic, stochastic, or...
In recent years, automatic classifiers of image cutouts (also called "stamps") have shown to be key for fast supernova discovery. The Vera C. Rubin Observatory will distribute about ten million alerts with their respective stamps each night, enabling the discovery approximately one supernovae year. A growing source confusion these is presence satellite glints, sequences point-like sources produced by rotating satellites or debris. currently planned a size smaller than typical separation...
Energy consumption represents a high operational cost in mining operations. Ore size reduction stage is the main consumer that process, where Semiautogenous Mill (SAG) one of components. The implementation control and automation strategies can achieve production goals along with energy efficiency are common goal concentrator plants; however, designing such controls requires proper understanding process dynamics which highly complex. This work studies machine learning deep be used to generate...
Energy consumption represents a significant operating expense in the mining and minerals industry. Grinding accounts for more than half of sector’s total energy usage, where semi-autogenous grinding (SAG) circuits are one main components. The implementation control automation strategies that can achieve production objectives along with efficiency is common goal concentrator plants. However, designing such controls requires proper understanding process dynamics, which highly complex, coupled,...
The advent of next-generation survey instruments, such as the Vera C. Rubin Observatory and its Legacy Survey Space Time (LSST), is opening a window for new research in time-domain astronomy. Extended LSST Astronomical Time-Series Classification Challenge (ELAsTiCC) was created to test capacity brokers deal with simulated stream. We describe ATAT, Transformer time series And Tabular data, classification model conceived by ALeRCE alert broker classify light-curves from streams. ATAT tested...
With a growing number of facilities able to monitor the entire sky and produce light curves with cadence days, in recent years there has been an increased rate detection sources whose variability deviates from standard behavior, revealing variety exotic nuclear transients. The aim present study is disentangle nature transient AT\,2021hdr, optical curve used be consistent classic Seyfert 1 nucleus, which was also confirmed by its spectrum high-energy properties. From late 2021, AT\,2021hdr...
A significant challenge in the study of transient astrophysical phenomena is identification bogus events, with human-made Earth-orbiting satellites and debris remain a key contaminant. Existing pipelines effectively identify satellite trails but can miss more complex signatures, such as collections dots known glints. In Rubin Observatory era, scale operations will increase tenfold respect to its precursor, Zwicky Transient Facility (ZTF), requiring crucial improvements classification purity,...