Transient classification in LIGO data using difference boosting neural network

Detection and classification of transients in data from gravitational wave detectors are crucial for efficient searches for true astrophysical events and identification of noise sources. We present a hybrid method for classification of short duration transients seen in gravitational wave data using both supervised and unsupervised machine learning techniques. To train the classifiers, we use the relative wavelet energy and the corresponding entropy obtained by applying one-dimensional wavelet decomposition on the data. The prediction accuracy of the trained classifier on nine simulated classes of gravitational wave transients and also LIGO’s sixth science run hardware injections are reported. Targeted searches for a couple of known classes of nonastrophysical signals in the first observational run of Advanced LIGO data are also presented. The ability to accurately identify transient classes using minimal training samples makes the proposed method a useful tool for LIGO detector characterization as well as searches for short duration gravitational wave signals.

Figure 1

Left panel depicts typical transient events (SNR set to 50 for better visualization). Wavelet energy median distribution for simulated data (SNR varied from 8 to 100) shown in the right panel.

Figure 2

Dendrogram showing hierarchical clustering of 1000 transient triggers identified in O1 Data from Hanford observatory by the Omicron algorithm [32]. The transient morphology changes progressively from left to right.

Figure 3

Block diagram of the proposed hybrid classifier.

Figure 4

Confusion matrix for simulated data: Results from traditional SVM (left) and DBNN (right) classifiers. The closer a diagonal element is to unity, the better is the classification for the corresponding type. Accuracy of our method is thus evident.

Figure 5

Donut chart shows the distribution of O1 transients classified by DBNN-Wavelet classifier into the major classes identified by hierarchical clustering. Representative transients from each class (sampled at 4096 Hz) are also shown.

Figure 6

OmegaScan: Type C glitch in LHO strain channel (left) and lightning glitch in LLO magnetometer (right). Plots generated through LIGOdv-web [45].