Skip to content
Myopic Posterior Sampling for Adaptive Bayesian Design of Experiments
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.


Myopic Posterior Sampling for Adaptive Goal Oriented Design of Experiments

MPS is a general and flexible framework adaptive goal oriented design of experiments.

In adaptive design of experiments (DoE), one wishes to design a sequence of experiments and collect data so as to achieve a desired goal. While there are many algorithms for specialised settings for adaptive DoE (such as optimisation, active learning, level set estimation etc.), MPS aims to provide a general framework that encompasses a broad variety of problems, including those mentioned above. To do so, one must specifiy their goal via a reward function. For more details, see our paper.

This library is compatible with Python2 (>= 2.7) and Python3 (>= 3.5) and has been tested on Linux and macOS platforms.


Installation & Getting Started

This library can be installed via the following commands.

$ git clone
$ cd mps
$ python install

Testing the installation: Once done, you may test the installation by importing mps in the python shell.

$ python
$ import mps

Getting started: To help get started, we have provided a few example scripts in the examples directory. Simply cd examples and run the script using python, e.g. python



Research and development of the methods in this package were funded by the Toyota Research Institute, Accelerated Materials Design & Discovery (AMDD) program.


If you use any part of this code in your work, please cite our ICML 2019 paper.

  title={Myopic Posterior Sampling for Adaptive Goal Oriented Design of Experiments},
  author={Kandasamy, Kirthevasan and Neiswanger, Willie and Zhang, Reed and Krishnamurthy,
Akshay and Schneider, Jeff and Poczos, Barnabas},
  booktitle={International Conference on Machine Learning},


This software is released under the MIT license. For more details, please refer LICENSE.txt.

For questions, please email

"Copyright 2018-2019 Kirthevasan Kandasamy"

You can’t perform that action at this time.