An official website of the United States government
Abstract Transient signals of instrumental and environmental origins (‘glitches’) in gravitational wave data elevate the false alarm rate of searches for astrophysical signals and reduce their sensitivity. Glitches that directly overlap astrophysical signals hinder their detection and worsen parameter estimation errors. As the fraction of data occupied by detectable astrophysical signals will be higher in next generation detectors, such problematic overlaps could become more frequent. These adverse effects of glitches can be mitigated by estimating and subtracting them out from the data, but their unpredictable waveforms and large morphological diversity pose a challenge. Subtraction of glitches using data from auxiliary sensors as predictors works but not for the majority of cases. Thus, there is a need for nonparametric glitch mitigation methods that do not require auxiliary data, work for a large variety of glitches, and have minimal effect on astrophysical signals in the case of overlaps. In order to cope with the high rate of glitches, it is also desirable that such methods be computationally fast. We show that adaptive spline fitting, in which the placement of free knots is optimized to estimate both smooth and non-smooth curves in noisy data, offers a promising approach to satisfying these requirements for broadband short-duration glitches, the type that appear quite frequently. The method is demonstrated on glitches drawn from three distinct classes in the Gravity Spy database as well as on the glitch that overlapped the binary neutron star signal GW170817. The impact of glitch subtraction on the GW170817 signal, or those like it injected into the data, is seen to be negligible.
more »
« less
Classification of time series as images using deep convolutional neural networks: application to glitches in gravitational wave data
The classification of frequently occurring terrestrial-origin transient signals, called glitches, in the time series data from gravitational wave detectors is important for mitigating their adverse effects on searches for rare and valuable astrophysical signals. While formally a time series classification problem, recent successes in glitch classification have all come from using their time-frequency image representations. Using transfer learning with the VGG16 deep convolutional neural network for image classification, we compare the efficacy of different types of image representations for classifying simulated glitches. We find the novel result that training the network with 2D plots of the noisy glitch time series provides better classification accuracy than their time-frequency images.
more »
« less
- Award ID(s):
- 2207935
- PAR ID:
- 10411304
- Date Published:
- Journal Name:
- 5th International Conference on Advances in Signal Processing and Artificial Intelligence (ASPAI' 2023)
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
The first successful detection of gravitational waves by ground-based observatories, such as the Laser Interferometer Gravitational-Wave Observatory (LIGO), marked a revolutionary breakthrough in our comprehension of the Universe. However, due to the unprecedented sensitivity required to make such observations, gravitational-wave detectors also capture disruptive noise sources called glitches, potentially masking or appearing as gravitational wave signals themselves. To address this problem, a community-science project, Gravity Spy, incorporates human insight and machine learning to classify glitches in LIGO data. The machine learning classifier, integrated into the project since 2017, has evolved over time to accommodate increasing numbers of glitch classes. Despite its success, limitations have arisen in the ongoing LIGO fourth observing run (O4) due to its architecture’s simplicity, which led to poor generalization and inability to handle multi-time window inputs effectively. We propose an advanced classifier for O4 glitches. Our contributions include evaluating fusion strategies for multi-time window inputs, using label smoothing to counter noisy labels, and enhancing interpretability through attention module-generated weights. This development seeks to enhance glitch classification, aiding in the ongoing exploration of gravitational-wave phenomena.more » « less
-
{"Abstract":["This dataset contains machine learning and volunteer classifications from the Gravity Spy project. It includes glitches from observing runs O1, O2, O3a and O3b that received at least one classification from a registered volunteer in the project. It also indicates glitches that are nominally retired from the project using our default set of retirement parameters, which are described below. See more details in the Gravity Spy Methods paper. <\/p>\n\nWhen a particular subject in a citizen science project (in this case, glitches from the LIGO datastream) is deemed to be classified sufficiently it is "retired" from the project. For the Gravity Spy project, retirement depends on a combination of both volunteer and machine learning classifications, and a number of parameterizations affect how quickly glitches get retired. For this dataset, we use a default set of retirement parameters, the most important of which are: <\/p>\n\nA glitches must be classified by at least 2 registered volunteers<\/li>Based on both the initial machine learning classification and volunteer classifications, the glitch has more than a 90% probability of residing in a particular class<\/li>Each volunteer classification (weighted by that volunteer's confusion matrix) contains a weight equal to the initial machine learning score when determining the final probability<\/li><\/ol>\n\nThe choice of these and other parameterization will affect the accuracy of the retired dataset as well as the number of glitches that are retired, and will be explored in detail in an upcoming publication (Zevin et al. in prep). <\/p>\n\nThe dataset can be read in using e.g. Pandas: \n```\nimport pandas as pd\ndataset = pd.read_hdf('retired_fulldata_min2_max50_ret0p9.hdf5', key='image_db')\n```\nEach row in the dataframe contains information about a particular glitch in the Gravity Spy dataset. <\/p>\n\nDescription of series in dataframe<\/strong><\/p>\n\n['1080Lines', '1400Ripples', 'Air_Compressor', 'Blip', 'Chirp', 'Extremely_Loud', 'Helix', 'Koi_Fish', 'Light_Modulation', 'Low_Frequency_Burst', 'Low_Frequency_Lines', 'No_Glitch', 'None_of_the_Above', 'Paired_Doves', 'Power_Line', 'Repeating_Blips', 'Scattered_Light', 'Scratchy', 'Tomte', 'Violin_Mode', 'Wandering_Line', 'Whistle']\n\tMachine learning scores for each glitch class in the trained model, which for a particular glitch will sum to unity<\/li><\/ul>\n\t<\/li>['ml_confidence', 'ml_label']\n\tHighest machine learning confidence score across all classes for a particular glitch, and the class associated with this score<\/li><\/ul>\n\t<\/li>['gravityspy_id', 'id']\n\tUnique identified for each glitch on the Zooniverse platform ('gravityspy_id') and in the Gravity Spy project ('id'), which can be used to link a particular glitch to the full Gravity Spy dataset (which contains GPS times among many other descriptors)<\/li><\/ul>\n\t<\/li>['retired']\n\tMarks whether the glitch is retired using our default set of retirement parameters (1=retired, 0=not retired)<\/li><\/ul>\n\t<\/li>['Nclassifications']\n\tThe total number of classifications performed by registered volunteers on this glitch<\/li><\/ul>\n\t<\/li>['final_score', 'final_label']\n\tThe final score (weighted combination of machine learning and volunteer classifications) and the most probable type of glitch<\/li><\/ul>\n\t<\/li>['tracks']\n\tArray of classification weights that were added to each glitch category due to each volunteer's classification<\/li><\/ul>\n\t<\/li><\/ul>\n\n <\/p>\n\n```\nFor machine learning classifications on all glitches in O1, O2, O3a, and O3b, please see Gravity Spy Machine Learning Classifications on Zenodo<\/p>\n\nFor the most recently uploaded training set used in Gravity Spy machine learning algorithms, please see Gravity Spy Training Set on Zenodo.<\/p>\n\nFor detailed information on the training set used for the original Gravity Spy machine learning paper, please see Machine learning for Gravity Spy: Glitch classification and dataset on Zenodo. <\/p>"]}more » « less
-
Abstract Gravitational wave (GW) interferometers are able to detect a change in distance of ~1/10 000th the size of a proton. Such sensitivity leads to large rates of non-gaussian, transient bursts of noise, also known as glitches, which hinder the detection and parameter estimation of short- and long-lived GW signals in the main detector strain. Glitches, come in a wide range of frequency-amplitude-time morphologies and may be caused by environmental or instrumental processes, so a key step towards their mitigation is to understand their population. Current approaches for their identification use supervised models to learn their morphology in the main strain with a fixed set of classes, but do not consider relevant information provided by auxiliary channels that monitor the state of the interferometers. In this work, we present an unsupervised algorithm to find anomalous glitches. Firstly, we encode a subset of auxiliary channels from Laser Interferometer Gravitational-Wave Observatory Livingston in the fractal dimension (FD), which measures the complexity of the signal. For this aim, we speed up the fractal dimension calculation to encode h of data in s. Secondly, we learn the underlying distribution of the data using an autoencoder with cyclic periodic convolutions. In this way, we learn the underlying distribution of glitches and we uncover unknown glitch morphologies, and overlaps in time between different glitches and misclassifications. This led to the discovery of anomalies in the input data. The results of this investigation stress the learnable structure of auxiliary channels encoded in FD and provide a flexible framework for glitch discovery.more » « less
-
ABSTRACT The global network of interferometric gravitational wave (GW) observatories (LIGO, Virgo, KAGRA) has detected and characterized nearly 100 mergers of binary compact objects. However, many more real GWs are lurking sub-threshold, which need to be sifted from terrestrial-origin noise triggers (known as glitches). Because glitches are not due to astrophysical phenomena, inference on the glitch under the assumption it has an astrophysical source (e.g. binary black hole coalescence) results in source parameters that are inconsistent with what is known about the astrophysical population. In this work, we show how one can extract unbiased population constraints from a catalogue of both real GW events and glitch contaminants by performing Bayesian inference on their source populations simultaneously. In this paper, we assume glitches come from a specific class with a well-characterized effective population (blip glitches). We also calculate posteriors on the probability of each event in the catalogue belonging to the astrophysical or glitch class, and obtain posteriors on the number of astrophysical events in the catalogue, finding it to be consistent with the actual number of events included.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
