Accelerating LISA inference with Gaussian processes

Abstract

Source inference for deterministic gravitational waves is a computationally demanding task in LISA. In a novel approach, we investigate the capability of Gaussian Processes to learn the posterior surface in order to reconstruct individual signal posteriors. We use GPry, which automates this reconstruction through active learning, using a very small number of likelihood evaluations, without the need for pretraining. We benchmark GPry against the cutting-edge nested sampler nessai, by injecting individually three signals on LISA noisy data simulated with Balrog: a white dwarf binary (DWD), a stellar-mass black hole binary (stBHB), and a super-massive black hole binary (SMBHB). We find that GPry needs $\mathcal{O}(10^{-2})$ fewer likelihood evaluations to achieve an inference accuracy comparable to nessai, with Jensen-Shannon divergence $D_{\mathrm{JS}}\lesssim 0.01$ for the DWD, and $D_{\mathrm{JS}}\lesssim 0.05$ for the SMBHB. Lower accuracy is found for the less Gaussian posterior of the stBHB: $D_{\mathrm{JS}}\lesssim 0.2$. Despite the overhead costs of GPry, we obtain a speed-up of $\mathcal{O}(10^2)$ for the slowest cases of stBHB and SMBHB. In conclusion, active-learning Gaussian process frameworks show great potential for rapid LISA parameter inference, especially for costly likelihoods, enabling suppression of computational costs without the trade-off of approximations in the calculations.

21 pages, 12 figures (Phys.Rev.D - Accepted 10 July, 2025)

References in corpus (39)

• SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python
• Array Programming with NumPy
• Search for an isotropic gravitational-wave background with the Parkes Pulsar Timing Array
• Searching for the nano-Hertz stochastic gravitational wave background with the Chinese Pulsar Timing Array Data Release I
• The second data release from the European Pulsar Timing Array III. Search for gravitational wave signals
• The NANOGrav 15-year Data Set: Search for Signals from New Physics
• Cobaya: Code for Bayesian Analysis of hierarchical physical models
• Computationally efficient models for the dominant and sub-dominant harmonic modes of precessing binary black holes
• Efficient sampling of fast and slow cosmological parameters
• PolyChord: nested sampling for cosmology
• PolyChord: next-generation nested sampling
• Cosmology with the Laser Interferometer Space Antenna
• IMRPhenomXHM: A multi-mode frequency-domain model for the gravitational wave signal from non-precessing black-hole binaries
• Real-time gravitational-wave science with neural posterior estimation
• Fast prediction and evaluation of gravitational waveforms using surrogate models
• Bayesian parameter estimation using conditional variational autoencoders for gravitational-wave astronomy
• Fast and Accurate Inference on Gravitational Waves from Precessing Compact Binaries
• Time-Delay Interferometry
• Enhancing Gravitational-Wave Science with Machine Learning
• Nested Sampling with Normalising Flows for Gravitational-Wave Inference
• Accelerated gravitational-wave parameter estimation with reduced order modeling
• Characterization of the stochastic signal originating from compact binaries populations as measured by LISA
• Neural Importance Sampling for Rapid and Reliable Gravitational-Wave Inference
• Massive black hole binaries in LISA: multimessenger prospects and electromagnetic counterparts
• Bayesian parameter estimation of stellar-mass black-hole binaries with LISA
• Real-time gravitational-wave inference for binary neutron stars using machine learning
• Identifying LISA verification binaries among the Galactic population of double white dwarfs
• Importance nested sampling with normalising flows
• Improved post-Newtonian waveform model for inspiralling precessing-eccentric compact binaries
• Gravitational Waves from Double White Dwarfs as probes of the Milky Way
• Chasing Super-Massive Black Hole merging events with $Athena$ and LISA
• Bayesian parameter-estimation of Galactic binaries in LISA data with Gaussian Process Regression
• Fast and robust Bayesian Inference using Gaussian Processes with GPry
• LISA stellar-mass black hole searches with semicoherent and particle-swarm methods
• Weighing Milky Way Satellites with LISA
• On the LISA science performance in observations of short-lived signals from massive black hole binary coalescences
• Stellar-mass black-hole binaries in LISA: characteristics and complementarity with current-generation interferometers
• A GPU-accelerated semi-coherent hierarchical search for stellar-mass binary inspiral signals in LISA
• Accelerating Bayesian Sampling for Massive Black Hole Binaries with Prior Constraints from Conditional Variational Autoencoder

Cited by in corpus (1)

• Accurate and efficient simulation-based inference for massive black-hole binaries with LISA