Flow Matching for Atmospheric Retrieval of Exoplanets: Where Reliability meets Adaptive Noise Levels
arXiv:2410.21477 · doi:10.1051/0004-6361/202451861
Abstract
Inferring atmospheric properties of exoplanets from observed spectra is key to understanding their formation, evolution, and habitability. Since traditional Bayesian approaches to atmospheric retrieval (e.g., nested sampling) are computationally expensive, a growing number of machine learning (ML) methods such as neural posterior estimation (NPE) have been proposed. We seek to make ML-based atmospheric retrieval (1) more reliable and accurate with verified results, and (2) more flexible with respect to the underlying neural networks and the choice of the assumed noise models. First, we adopt flow matching posterior estimation (FMPE) as a new ML approach to atmospheric retrieval. FMPE maintains many advantages of NPE, but provides greater architectural flexibility and scalability. Second, we use importance sampling (IS) to verify and correct ML results, and to compute an estimate of the Bayesian evidence. Third, we condition our ML models on the assumed noise level of a spectrum (i.e., error bars), thus making them adaptable to different noise models. Both our noise level-conditional FMPE and NPE models perform on par with nested sampling across a range of noise levels when tested on simulated data. FMPE trains about 3 times faster than NPE and yields higher IS efficiencies. IS successfully corrects inaccurate ML results, identifies model failures via low efficiencies, and provides accurate estimates of the Bayesian evidence. FMPE is a powerful alternative to NPE for fast, amortized, and parallelizable atmospheric retrieval. IS can verify results, thus helping to build confidence in ML-based approaches, while also facilitating model comparison via the evidence ratio. Noise level conditioning allows design studies for future instruments to be scaled up, for example, in terms of the range of signal-to-noise ratios.
Accepted for publication in Astronomy & Astrophysics
References in corpus (20)
- Array Programming with NumPy
- petitRADTRANS: a Python radiative transfer package for exoplanet characterization and retrieval
- Real-time gravitational-wave science with neural posterior estimation
- Retrieving scattering clouds and disequilibrium chemistry in the atmosphere of HR 8799e
- Neural Importance Sampling for Rapid and Reliable Gravitational-Wave Inference
- An Ensemble of Bayesian Neural Networks for Exoplanetary Atmospheric Retrieval
- Large Interferometer For Exoplanets (LIFE): III. Spectral resolution, wavelength range and sensitivity requirements based on atmospheric retrieval analyses of an exo-Earth
- Four-of-a-kind? Comprehensive atmospheric characterisation of the HR 8799 planets with VLTI/GRAVITY
- Atmospheric retrievals with petitRADTRANS
- Neural posterior estimation for exoplanetary atmospheric retrieval
- ExoMDN: Rapid characterization of exoplanet interior structures with Mixture Density Networks
- normflows: A PyTorch Package for Normalizing Flows
- Assessment of Supervised Machine Learning for Atmospheric Retrieval of Exoplanets
- Convolutional neural networks as an alternative to Bayesian retrievals
- Adapting to noise distribution shifts in flow-based gravitational-wave inference
- Large Interferometer For Exoplanets (LIFE): XIII. The Value of Combining Thermal Emission and Reflected Light for the Characterization of Earth Twins
- Notes on the Practical Application of Nested Sampling: MultiNest, (Non)convergence, and Rectification
- Impacts of high-contrast image processing on atmospheric retrievals
- Parameterizing pressure-temperature profiles of exoplanet atmospheres with neural networks
- Using a neural network approach to accelerate disequilibrium chemistry calculations in exoplanet atmospheres