Identification of problematic epochs in astronomical time series through transfer learning
arXiv:2405.05591 · doi:10.1051/0004-6361/202450166
Abstract
We present a novel method for detecting outliers in astronomical time series based on the combination of a deep neural network and a k-nearest neighbor algorithm with the aim of identifying and removing problematic epochs in the light curves of astronomical objects. We use an EfficientNet network pre-trained on ImageNet as a feature extractor and perform a k-nearest neighbor search in the resulting feature space to measure the distance from the first neighbor for each image. If the distance is above the one obtained for a stacked image, we flag the image as a potential outlier. We apply our method to time series obtained from the VLT Survey Telescope (VST) monitoring campaign of the Deep Drilling Fields of the Vera C. Rubin Legacy Survey of Space and Time (LSST). We show that our method can effectively identify and remove artifacts from the VST time series and improve the quality and reliability of the data. This approach may prove very useful in sight of the amount of data that will be provided by the LSST, which will prevent the inspection of individual light curves. We also discuss the advantages and limitations of our method and suggest possible directions for future work.
arXiv commentaccepted for publication in A&A
References in corpus (14)
- LSST: from Science Drivers to Reference Design and Anticipated Data Products
- Supernova rates from the SUDARE VST-Omegacam search. I
- Variability-selected Active Galactic Nuclei in the VST-SUDARE/VOICE Survey of the COSMOS Field
- Supernova rates from the SUDARE VST-Omegacam search II. Rates in a galaxy sample
- Optically variable active galactic nuclei in the 3 yr VST survey of the COSMOS field
- A structure function analysis of VST-COSMOS AGN
- SUDARE-VOICE variability-selection of Active Galaxies in the Chandra Deep Field South and the SERVS/SWIRE region
- A random forest-based selection of optically variable AGN in the VST-COSMOS field
- The Photometric LSST Astronomical Time-series Classification Challenge (PLAsTiCC): Selection of a performance metric for classification probabilities balancing diverse science goals
- Classification of Planetary Nebulae through Deep Transfer Learning
- Extending the variability selection of active galactic nuclei in the W-CDF-S and SERVS/SWIRE region
- Weak Lensing Study in VOICE Survey I: Shear Measurement
- ULISSE: A Tool for One-shot Sky Exploration and its Application to Active Galactic Nuclei Detection
- Variability and transient search in the SUDARE-VOICE field: a new method to extract the light curves