ALABI: Active Learning for Accelerated Bayesian Inference

We present Active Learning for Accelerated Bayesian Inference ($\mathtt{\text{alabi}}$): an open-source Python package for performing Bayesian inference with computationally expensive models. Given a forward model and observational data to construct a likelihood and priors, $\mathtt{\text{alabi}}$\ uses a Gaussian Process (GP) surrogate model trained to predict posterior probability as a function of input parameters, and employs active learning to iteratively improve GP predictive performance in high-likelihood regions where the GP is most uncertain. $\mathtt{\text{alabi}}$\ provides a uniform interface for using Markov chain Monte Carlo (MCMC) with different packages, including the affine-invariant sampler $\mathtt{\text{emcee}}$, and nested samplers $\mathtt{\text{dynesty}}$, $\mathtt{\text{multinest}}$, and $\mathtt{\text{ultranest}}$. This approach facilitates accurate estimation of the desired posterior distribution, while reducing the number of computationally expensive model evaluations required by factors of thousands. We demonstrate the performance of $\mathtt{\text{alabi}}$\ on a variety of test cases, including where inference is challenging due to complex posterior structure or high dimensionality. We show that $\mathtt{\text{alabi}}$\ offers a substantial improvement for likelihood functions with evaluation times $\gtrsim 1$,s, speeding up MCMC computations by a factor of $10-1000\times$ when tested on problems with up to 64 dimensions.