Real-Time Likelihood-Free Inference of Roman Binary Microlensing Events with Amortized Neural Posterior Estimation

Abstract

Fast and automated inference of binary-lens, single-source (2L1S) microlensing events with sampling-based Bayesian algorithms (e.g., Markov Chain Monte Carlo; MCMC) is challenged on two fronts: high computational cost of likelihood evaluations with microlensing simulation codes, and a pathological parameter space where the negative-log-likelihood surface can contain a multitude of local minima that are narrow and deep. Analysis of 2L1S events usually involves grid searches over some parameters to locate approximate solutions as a prerequisite to posterior sampling, an expensive process that often requires human-in-the-loop domain expertise. As the next-generation, space-based microlensing survey with the Roman Space Telescope is expected to yield thousands of binary microlensing events, a new fast and automated method is desirable. Here, we present a likelihood-free inference (LFI) approach named amortized neural posterior estimation, where a neural density estimator (NDE) learns a surrogate posterior $\hat{p}(\theta |x)$ as an observation-parametrized conditional probability distribution, from pre-computed simulations over the full prior space. Trained on 291,012 simulated Roman-like 2L1S simulations, the NDE produces accurate and precise posteriors within seconds for any observation within the prior support without requiring a domain expert in the loop, thus allowing for real-time and automated inference. We show that the NDE also captures expected posterior degeneracies. The NDE posterior could then be refined into the exact posterior with a downstream MCMC sampler with minimal burn-in steps.

15 pages, 8 figures, 3 tables. Accepted for publication in the Astronomical Journal. This article supersedes arXiv:2010.04156

References in corpus (21)

• The NumPy array: a structure for efficient numerical computation
• WaveNet: A Generative Model for Raw Audio
• SGDR: Stochastic Gradient Descent with Warm Restarts
• The frontier of simulation-based inference
• Wide-Field InfrarRed Survey Telescope-Astrophysics Focused Telescope Assets WFIRST-AFTA 2015 Report
• The Use of High Magnification Microlensing Events in Discovering Extra-solar Planets
• No large population of unbound or wide-orbit Jupiter-mass planets
• Predictions of the WFIRST Microlensing Survey I: Bound Planet Detection Rates
• Planet Parameters in Microlensing Events
• A recurrent neural network for classification of unevenly sampled variable stars
• Microlensing of Blended Stellar Images
• OGLE-III Microlensing Events and the Structure of the Galactic Bulge
• Sequential Neural Likelihood: Fast Likelihood-free Inference with Autoregressive Flows
• Modeling microlensing events with MulensModel
• A New Type of Ambiguity in the Planet and Binary Interpretations of Central Perturbations of High-Magnification Gravitational Microlensing Events
• Distinguishing between Planetary and Binary Interpretations of Microlensing Central Perturbations under Severe Finite-Source Effect
• A Machine Learning Classifier for Microlensing in Wide-Field Surveys
• Ogle-2018-blg-0677lb: A super earth near the galactic bulge
• Classification of Periodic Variable Stars with Novel Cyclic-Permutation Invariant Neural Networks
• A Fast Approximate Approach to Microlensing Survey Analysis
• Automating Inference of Binary Microlensing Events with Neural Density Estimation

Cited by in corpus (13)

• Accelerated Bayesian SED Modeling using Amortized Neural Posterior Estimation
• A Ubiquitous Unifying Degeneracy in Two-Body Microlensing Systems
• $\mathrm{S}\mathrm{IM}\mathrm{BIG}$: Mock Challenge for a Forward Modeling Approach to Galaxy Clustering
• Fast and robust Bayesian Inference using Gaussian Processes with GPry
• An Earth-Mass Planet and a Brown Dwarf in Orbit Around a White Dwarf
• CoLFI: Cosmological Likelihood-free Inference with Neural Density Estimators
• A microlensing search of 700 million VVV light curves
• PopSED: Population-Level Inference for Galaxy Properties from Broadband Photometry with Neural Density Estimation
• DIGS: Deep Inference of Galaxy Spectra with Neural Posterior Estimation
• 60 Microlensing Events from the Three Years of Zwicky Transient Facility Phase One
• MAGIC: Microlensing Analysis Guided by Intelligent Computation
• A differentiable binary microlensing model using adaptive contour integration method
• Binary-lens Microlensing Degeneracy: Impact on Planetary Sensitivity and Mass-ratio Function