1: \begin{abstract}

2: Calculation of Bayesian posteriors and model evidences typically requires numerical integration.

3: Bayesian quadrature (BQ), a surrogate-model-based approach to numerical integration, is capable of superb sample efficiency, but its lack of parallelisation has hindered its practical applications.

4: In this work, we propose a parallelised (batch) BQ method, employing techniques from kernel quadrature, that possesses an empirically exponential convergence rate.

5: Additionally, just as with Nested Sampling, our method permits simultaneous inference of both posteriors and model evidence.

6: Samples from our BQ surrogate model are re-selected to give a sparse set of samples, via a kernel recombination algorithm, requiring negligible additional time to increase the batch size.

7: Empirically, we find that our approach significantly outperforms the sampling efficiency of both state-of-the-art BQ techniques and Nested Sampling in various real-world datasets, including lithium-ion battery analytics.\footnote{Code: \url{https://github.com/ma921/BASQ}}\looseness=-1

8: \end{abstract}

9: