Changelog

All notable changes to BOBE will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[0.1.0] - 2025-12-XX

Initial release of BOBE (Bayesian Optimization for Bayesian Evidence).

Added

  • Bayesian optimization for efficient Bayesian evidence estimation

  • JAX-powered Gaussian Process surrogates with RBF and Matérn kernels

  • GP priors: DSLP (Deep Sigmoidal Location Process) and SAAS (Sparse Axis-Aligned Subspace)

  • Classifier-augmented GP (SVM, neural network, ellipsoid) for filtering low-likelihood regions

  • Acquisition functions: WIPV, WIPStd, EI, LogEI

  • Nested sampling via Dynesty for evidence computation on GP surrogate

  • Cobaya integration for cosmological likelihoods

  • MPI parallelization for likelihood evaluations and GP fitting

  • Result saving and visualization utilities

  • Convergence monitoring via KL divergence and log-evidence thresholds