Bayesian Adaptive Direct Search (BADS) optimizer

Bayesian Adaptive Direct Search (BADS) optimization algorithm for model fitting

BADS is a novel, fast Bayesian optimization algorithm designed to solve difficult optimization problems, in particular related to fitting computational models (e.g., via maximum likelihood estimation). In our benchmarks with real model-fitting problems, BADS performed on par or better than many other common and state-of-the-art MATLAB optimizers, such as fminsearch, fmincon, and cmaes [1].

BADS is currently being used in many computational labs around the world, with more than a hundred citations and applications ranging from behavioral, cognitive, and computational neuroscience to engineering and economics.

BADS is recommended when no gradient information is available, and the objective function is non-analytical or noisy, for example evaluated through numerical approximation or via simulation. BADS requires no specific tuning and runs off-the-shelf like other built-in MATLAB optimizers such as fminsearch.

*** For extensive information, tutorials and documentation, please visit the GitHub page of the project: https://github.com/lacerbi/bads ***

If you are interested in estimating posterior distributions (i.e., uncertainty and error bars) over parameters, and not just point estimates, you might want to check out Variational Bayesian Monte Carlo (VBMC), a toolbox for Bayesian posterior and model inference which can be used in synergy with BADS: https://github.com/lacerbi/vbmc