Learning Bayesian posteriors with neural networks for gravitational-wave inference
This work addresses the need for low-latency parameter estimation in gravitational-wave astronomy, offering a domain-specific solution that is incremental by building on existing neural-network methods.
The paper tackles the problem of performing Bayesian inference for gravitational-wave astronomy by using deep-learning techniques to instantly produce posterior distributions for source parameters from detector data, achieving efficient inference through a neural network trained on simulated data with a compact representation.
We seek to achieve the Holy Grail of Bayesian inference for gravitational-wave astronomy: using deep-learning techniques to instantly produce the posterior $p(θ|D)$ for the source parameters $θ$, given the detector data $D$. To do so, we train a deep neural network to take as input a signal + noise data set (drawn from the astrophysical source-parameter prior and the sampling distribution of detector noise), and to output a parametrized approximation of the corresponding posterior. We rely on a compact representation of the data based on reduced-order modeling, which we generate efficiently using a separate neural-network waveform interpolant [A. J. K. Chua, C. R. Galley & M. Vallisneri, Phys. Rev. Lett. 122, 211101 (2019)]. Our scheme has broad relevance to gravitational-wave applications such as low-latency parameter estimation and characterizing the science returns of future experiments. Source code and trained networks are available online at https://github.com/vallis/truebayes.