$\beta$-Annealed Variational Autoencoder for glitches

Abstract

Gravitational wave detectors such as LIGO and Virgo are susceptible to various types of instrumental and environmental disturbances known as glitches which can mask and mimic gravitational waves. While there are 22 classes of non-Gaussian noise gradients currently identified, the number of classes is likely to increase as these detectors go through commissioning between observation runs. Since identification and labelling new noise gradients can be arduous and time-consuming, we propose $\beta$-Annelead VAEs to learn representations from spectograms in an unsupervised way. Using the same formulation as \cite{alemi2017fixing}, we view Bottleneck-VAEs~cite{burgess2018understanding} through the lens of information theory and connect them to $\beta$-VAEs~cite{higgins2017beta}. Motivated by this connection, we propose an annealing schedule for the hyperparameter $\beta$ in $\beta$-VAEs which has advantages of: 1) One fewer hyperparameter to tune, 2) Better reconstruction quality, while producing similar levels of disentanglement.

References in corpus (8)

• Auto-Encoding Variational Bayes
• Advanced Virgo: a 2nd generation interferometric gravitational wave detector
• Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations
• Characterization of transient noise in Advanced LIGO relevant to gravitational wave signal GW150914
• Gravity Spy: Integrating Advanced LIGO Detector Characterization, Machine Learning, and Citizen Science
• Understanding disentangling in $\beta$-VAE
• Calibration of the Advanced LIGO detectors for the discovery of the binary black-hole merger GW150914
• Efficient Gravitational-wave Glitch Identification from Environmental Data Through Machine Learning

Cited by in corpus (1)

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