Simulation-Efficient Cosmological Inference with Multi-Fidelity SBI

The simulation cost for cosmological simulation-based inference can be decreased by combining simulation sets of varying fidelity. We propose an approach to such multi-fidelity inference based on feature matching and knowledge distillation. Our method results in improved posterior quality, particularly for small simulation budgets and difficult inference problems.

@article{thiele2025_2507.00514,
  title={ Simulation-Efficient Cosmological Inference with Multi-Fidelity SBI },
  author={ Leander Thiele and Adrian E. Bayer and Naoya Takeishi },
  journal={arXiv preprint arXiv:2507.00514},
  year={ 2025 }
}