Search Results for "bayesian optimization"

Nevergrad is a Python library for derivative-free optimization, offering robust implementations of many algorithms suited for black-box functions (i.e. functions where gradients are unavailable or unreliable). It targets hyperparameter search, architecture search, control problems, and experimental tuning—domains in which gradient-based methods may fail or be inapplicable. The library provides an easy interface to define an optimization problem (parameter space, loss function, budget) and then experiment with multiple strategies—evolutionary algorithms, Bayesian optimization, bandit methods, genetic algorithms, etc. ...

BayesianOptimization is a Python library that helps find the maximum (or minimum) of expensive or unknown objective functions using Bayesian optimization. This technique is especially useful for hyperparameter tuning in machine learning, where evaluating the objective function is costly. The library provides an easy-to-use API for defining bounds and optimizing over parameter spaces using probabilistic models like Gaussian Processes.

Meridian is a comprehensive, open source marketing mix modeling (MMM) framework developed by Google to help advertisers analyze and optimize the impact of their marketing investments. Built on Bayesian causal inference principles, Meridian enables organizations to evaluate how different marketing channels influence key performance indicators (KPIs) such as revenue or conversions while accounting for external factors like seasonality or economic trends. The framework provides a robust...

...PyMC3 provides rich support for defining and using GPs. Variational inference saves computational cost by turning a problem of integration into one of optimization. PyMC3's variational API supports a number of cutting edge algorithms, as well as minibatch for scaling to large datasets.

RStan is the R interface to Stan, a C++ library for statistical modeling and high-performance statistical computation. It lets users specify models in the Stan modeling language (for Bayesian inference), compile them, and perform inference from R. Key inference approaches include full Bayesian inference via Hamiltonian Monte Carlo (specifically the No-U-Turn Sampler, NUTS), approximate Bayesian inference via variational methods, and optimization (penalized likelihood). RStan integrates with Stan’s automatic differentiation library, provides diagnostics, model comparison, posterior predictive checks, etc. ...