Efficient gravitational-wave glitch identification from environmental data through machine learning

@article{Colgan2020EfficientGG,
  title={Efficient gravitational-wave glitch identification from environmental data through machine learning},
  author={Robert E. Colgan and K. Rainer Corley and Yenson Lau and Imre Bartos and John N. Wright and Zsuzsanna Marka and Szabolcs M{\'a}rka},
  journal={Physical Review D},
  year={2020},
  volume={101},
  pages={102003}
}
The LIGO observatories detect gravitational waves through monitoring changes in the detectors' length down to below ${10}^{\ensuremath{-}19}\text{ }\text{ }\mathrm{m}/\sqrt{\mathrm{Hz}}$ variations---a small fraction of the size of the atoms that make up the detector. To achieve this sensitivity, the detector and its environment need to be closely monitored. Beyond the gravitational-wave data stream, LIGO continuously records hundreds of thousands of channels of environmental and instrumental… 

Architectural Optimization and Feature Learning for High-Dimensional Time Series Datasets

Robert E. Colgan, Jingkai Yan, Zsuzsa MáRka, Imre Bartos, Szabolcs Márka, and John N. Wright's dissertation aims to provide a history of particle physics in the Large Hadron Collider from 1989 to 2002.

Boosting the Efficiency of Parametric Detection with Hierarchical Neural Networks

This work proposes Hierarchical Detection Network (HDN), a novel approach to efficient detection that combines ideas from hierarchical matching and deep learning, and shows how training a three-layer HDN initialized using two-layer model can further boost both accuracy and e-ciency.

Nonlinear Noise Cleaning in Gravitational-Wave Detectors With Convolutional Neural Networks

This study examines the requirements in the signal-to-noise ratio (SNR) in both the target channel and in the auxiliary sensors in order to reduce the noise by at least a factor of a few, and shows that the CNN can still reach a good performance if it is used in the case of limited SNR.

DeepAdversaries: examining the robustness of deep learning models for galaxy morphology classification

It is shown that training with domain adaptation improves model robustness and mitigates the effects of these perturbations, improving the classification accuracy up to 23% on data with higher observational noise and increasing the latent space distance between the baseline and the incorrectly classified one-pixel perturbed image.

Source-agnostic gravitational-wave detection with recurrent autoencoders

An application of anomaly detection techniques based on deep recurrent autoencoders (AEs) to the problem of detecting gravitational wave signals in laser interferometers and the recurrent AE outperforms other AEs based on different architectures.

Nonlinear noise regression in gravitational-wave detectors with convolutional neural networks

This study examines the requirement in the signal-to-noise ratio (SNR) in both the main GW readout and in the auxiliary sensors in order to reduce the noise by at least a factor of a few, and demonstrates that the CNN can still reach a good performance if it adopts curriculum learning techniques.

$\beta$-Annealed Variational Autoencoder for glitches

An annealing schedule for thehyperparameter β in β-VAEs is proposed which has advantages of: 1) One fewer hyperparameter to tune, 2) Better reconstruction quality, while producing similar levels of disentanglement.

On the Efficiency of Various Deep Transfer Learning Models in Glitch Waveform Detection in Gravitational-Wave Data

Even though the models achieved fairly high accuracy, it is observed that all of the model suffered from the lack of data for certain classes which is the main concern based on the results of the study.

A Deep Transfer Learning Approach on Identifying Glitch Wave-form in Gravitational Wave Data

It was observed that less complex algorithm like VGG19, DenseNet169 and VGG16 performs better than most of the more complex algorithms featured in this study which could possibly indicate that less more complex models might be preferred when identifying glitch wave-forms.

Gravitational-wave parameter inference using Deep Learning

It is shown that a classifier network can be trained in order to detect the presence of GW signal with high accuracy, and a regression network is trained to perform parameter inference on BBH spectrogram data.

References

SHOWING 1-10 OF 64 REFERENCES

Detectability of dynamical tidal effects and the detection of gravitational-wave transients with LIGO

Dynamical tidal effects impact the orbital motion of extended bodies, imprinting themselves in several measurable ways. This thesis explores the saturation of weakly nonlinear dynamical tidal

Pattern Recognition and Machine Learning (Springer-Verlag, Berlin

  • 2006

Modelos de mezclas Bernoulli con regresión logística: una aplicación en la valoración de carteras de crédito

Este trabajo final de maestria, modalidad de profundizacion, consiste en la elaboracion de un problema de modelacion estadistica aplicada al sector crediticio. El objetivo es aplicar un modelo de

Advanced Virgo: a 2nd generation interferometric gravitational wave detector

Advanced Virgo is the project to upgrade the Virgo interferometric detector of gravitational waves, with the aim of increasing the number of observable galaxies (and thus the detection rate) by three

Predicción y selección de variables con bosques aleatorios en presencia de variables correlacionadas

El presente trabajo aborda el problema de seleccion de variables empleando el metodo de bosques aleatorios cuando el modelo subyacente para la variable respuesta es de tipo lineal. Para ello se

Learning Bayesian Posteriors with Neural Networks for Gravitational-Wave Inference.

Using deep-learning techniques to instantly produce the posterior p(θ|D) for the source parameters θ, given the detector data D, has broad relevance to gravitational-wave applications such as low-latency parameter estimation and characterizing the science returns of future experiments.

Deep-learning continuous gravitational waves

We present a first proof-of-principle study for using deep neural networks (DNNs) as a novel search method for continuous gravitational waves (CWs) from unknown spinning neutron stars. The

Regularization and variable selection via the elastic net

It is shown that the elastic net often outperforms the lasso, while enjoying a similar sparsity of representation, and an algorithm called LARS‐EN is proposed for computing elastic net regularization paths efficiently, much like algorithm LARS does for the lamba.
...