Simulation-efficient marginal posterior estimation with swyft: stop wasting your precious time
This work addresses the problem of inefficient simulator usage for researchers in physics and astronomy, particularly for likelihood-free inference scenarios.
This paper introduces algorithms for nested neural likelihood-to-evidence ratio estimation and simulation reuse through a Poisson point process cache. These methods enable automatic and highly efficient estimation of marginal and joint posteriors, achieving an order of magnitude better simulator efficiency than traditional likelihood-based sampling.
We present algorithms (a) for nested neural likelihood-to-evidence ratio estimation, and (b) for simulation reuse via an inhomogeneous Poisson point process cache of parameters and corresponding simulations. Together, these algorithms enable automatic and extremely simulator efficient estimation of marginal and joint posteriors. The algorithms are applicable to a wide range of physics and astronomy problems and typically offer an order of magnitude better simulator efficiency than traditional likelihood-based sampling methods. Our approach is an example of likelihood-free inference, thus it is also applicable to simulators which do not offer a tractable likelihood function. Simulator runs are never rejected and can be automatically reused in future analysis. As functional prototype implementation we provide the open-source software package swyft.