Saul Toscano-Palmerin

Ph.D. Student, Cornell University

Bayesian Global Optimization Library (BGO)

BGO is a Bayesian Global Optimization framework written in Python by Saul Toscano-Palmerin. The library includes four different algorithms: Stratified Bayesian Optimization (SBO), Knowledge Gradient (KG), Expected Improvement (EI) and Probability Improvement (PI). Please refer to the Documentation.

This library is still in development and soon we are going to include examples where we need to decide how to optimally allocate computational/experimental effort across information sources, to optimize functions.

For one example optimized using this library, see CitiBike and Description of the Citi Bike Simulator.