A Strong Gravitational Lens Is Worth a Thousand Dark Matter Halos: Inference on Small-scale Structure Using Sequential Methods

Journal Article · 07 November 2024 · The Astrophysical Journal

DOI: https://doi.org/10.3847/1538-4357/ad6e70 · OSTI ID:2479069

Wagner-Carena, Sebastian  ^[1]; Lee, Jaehoon ^[2]; Pennington, Jeffrey ^[2]; Aalbers, Jelle ^[3]; Birrer, Simon  ^[4]; Wechsler, Risa H.  ^[5]

---

1. New York Univ. (NYU), NY (United States); Flatiron Institute, New York, NY (United States); Stanford Univ., CA (United States). Kavli Institute for Particle Astrophysics & Cosmology; Google DeepMind, London (United Kingdom)
2. Google DeepMind, London (United Kingdom)
3. Univ. of Groningen (Netherlands). Van Swinderen Institute for Particle Physics and Gravity; Stanford Univ., CA (United States). Kavli Institute for Particle Astrophysics & Cosmology
4. Stony Brook Univ., NY (United States)
5. Stanford Univ., CA (United States). Kavli Institute for Particle Astrophysics & Cosmology

Strong gravitational lenses are a singular probe of the Universe’s small-scale structure—they are sensitive to the gravitational effects of low-mass (<10¹⁰ M_⊙) halos even without a luminous counterpart. Recent strong-lensing analyses of dark matter structure rely on simulation-based inference (SBI). Modern SBI methods, which leverage neural networks as density estimators, have shown promise in extracting the halo-population signal. However, it is unclear whether the constraints from these models are limited by the methodology or the data. In this study, we introduce an accelerator-optimized simulation pipeline that can generate lens images with realistic subhalo populations in milliseconds. Leveraging this simulator, we identify the main limitation of our fiducial SBI analysis: training set size. We then adopt a sequential neural posterior estimation (SNPE) approach, allowing us to refine the training distribution to align with the observed data. Using only one-fifth as many mock Hubble Space Telescope images, SNPE matches the constraints on the low-mass halo population produced by our best nonsequential model. Our experiments suggest that an over 3 order-of-magnitude increase in training set size and GPU hours would be required to achieve an equivalent result without sequential methods. While the full potential of the existing lens sample remains to be explored, the notable improvement in constraining power enabled by our sequential approach highlights that current constraints are limited primarily by methodology and not the data itself. Moreover, our results emphasize the need to treat training set generation and model optimization as interconnected stages of any cosmological analysis using SBI.

• Research Organization: SLAC National Accelerator Laboratory (SLAC), Menlo Park, CA (United States)
• Sponsoring Organization: USDOE; National Aeronautics and Space Administration (NASA); National Science Foundation (NSF)
• Grant/Contract Number: AC02-76SF00515
• OSTI ID: 2479069
• Journal Information: The Astrophysical Journal, Journal Name: The Astrophysical Journal Journal Issue: 2 Vol. 975; ISSN 0004-637X
• Publisher: IOP PublishingCopyright Statement
• Country of Publication: United States
• Language: English