Speaker
Description
Gravitational-Wave detections employ the Matched Filtering based algorithms for making the detections. Matched Filtering works by matching data from the incoming data stream with waveform templates from the templates banks generated in some interested parameter space. Since the parameter space can be large if we want to include most of the detections, generating a template bank can involve high computational and time resources. If we have some Machine Learning based algorithm to generate the gravitational waveform templates, this complexity can be reduced by a large amount. In my talk and poster, I will introduce an Auto-Encoder model for this purpose and show how useful a future on-line implementation of it on the LIGO-Virgo-KAGRA data analysis pipeline can be for detections.
