Optbayesexpt: Sequential Bayesian experiment design for adaptive measurements

Abstract

Optbayesexpt is a public domain, open-source python package that provides adaptive algorithms for effcient estimation/measurement of parameters in a model function. Parameter estimation is the type of measurement one would conventionally tackle with a sequence of data acquisition steps followed by ftting. The software is designed to provide data-based control of experiments, effectively learning from incoming measurement results and using that information to select future measurement settings live and online as measurements progress. The settings are chosen to have the best chances of improving the measurement results. With these methods optbayesexpt is designed to increase the effciency of a sequence of measurements, yielding better results and/or lower cost. In a recent experiment, optbayesexpt yielded an order of magnitude increase in speed for measurement of a few narrow peaks in a broad spectral range [1]. Figure 1 illustrates a possible use scenario where an instrument control program interacts with optbayesexpt functions in a python server script. The server script runs in the background, and the two programs communicate via TCP sockets. For each of N measurements, the controller program prompts the server script to recommend high-utility settings based on the parameter probability distribution. The controller then makes measurements and reports the new data back to the server script, which uses Bayesian inference to refne the parameter distribution for use in the next iteration.