We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

astro-ph.IM

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Astrophysics > Instrumentation and Methods for Astrophysics

Title: Gravitational-wave parameter estimation with autoregressive neural network flows

Abstract: We introduce the use of autoregressive normalizing flows for rapid likelihood-free inference of binary black hole system parameters from gravitational-wave data with deep neural networks. A normalizing flow is an invertible mapping on a sample space that can be used to induce a transformation from a simple probability distribution to a more complex one: if the simple distribution can be rapidly sampled and its density evaluated, then so can the complex distribution. Our first application to gravitational waves uses an autoregressive flow, conditioned on detector strain data, to map a multivariate standard normal distribution into the posterior distribution over system parameters. We train the model on artificial strain data consisting of IMRPhenomPv2 waveforms drawn from a five-parameter $(m_1, m_2, \phi_0, t_c, d_L)$ prior and stationary Gaussian noise realizations with a fixed power spectral density. This gives performance comparable to current best deep-learning approaches to gravitational-wave parameter estimation. We then build a more powerful latent variable model by incorporating autoregressive flows within the variational autoencoder framework. This model has performance comparable to Markov chain Monte Carlo and, in particular, successfully models the multimodal $\phi_0$ posterior. Finally, we train the autoregressive latent variable model on an expanded parameter space, including also aligned spins $(\chi_{1z}, \chi_{2z})$ and binary inclination $\theta_{JN}$, and show that all parameters and degeneracies are well-recovered. In all cases, sampling is extremely fast, requiring less than two seconds to draw $10^4$ posterior samples.
Comments: 14 pages, 7 figures
Subjects: Instrumentation and Methods for Astrophysics (astro-ph.IM); Machine Learning (cs.LG); General Relativity and Quantum Cosmology (gr-qc); Machine Learning (stat.ML)
Journal reference: Phys. Rev. D 102, 104057 (2020)
DOI: 10.1103/PhysRevD.102.104057
Report number: LIGO-P2000053
Cite as: arXiv:2002.07656 [astro-ph.IM]
  (or arXiv:2002.07656v1 [astro-ph.IM] for this version)

Submission history

From: Stephen Green [view email]
[v1] Tue, 18 Feb 2020 15:44:04 GMT (2855kb,D)

Link back to: arXiv, form interface, contact.