Last modified 5 months ago Last modified on 05/11/17 07:55:35

Modified Thu May 11 07:55:35 2017 by Eric.Woldridge.

2017 PPAML Summer School Announcement

July 24th to August 4th


Are you a graduate student in the sciences, a developer or engineer interested in machine learning, or a statistician or data scientist?  Do you need to rapidly develop and test models or scientific theories of a complex system involving neurons, molecular biology, economics, social networks, astrophysics, etc.?  Or do you need to prepare data for analysis and/or bring together data from multiple sources, where some of data is paragraphs or blobs of text? 

As part of the Probabilistic Programming for Advancing Machine Learning (PPAML) program, Galois is hosting the fourth of four annual summer schools on the topic this July in Arlington, Virginia.

The two-week program is designed to teach participants the necessary background on probabilistic programming languages being developed as part of PPAML, and give them an opportunity to work directly with the creators of these languages to solve their own problems using these tools.

During the summer school, participants will provide feedback about the functionality, usability and performance of these probabilistic programming systems to help the PPAML program development teams improve their tools.

Languages and Tools

In the first week, the MIT team will feature:

  • AI-assisted data analysis with BayesDB --- learning how to answer data analysis questions in minutes that normally take hours or days for experienced computational statisticians.
  • Interactive probabilistic reasoning with Venture --- learning how to extend BayesDB analyses with causal knowledge, and to solve hard model selection and structure learning problems.

In the second week, the Gamalon team will feature the Particle platform for rapidly developing, running inference, and debugging/profiling models of complex systems of all kinds.  Participants will also learn about Gamalon’s Structure platform for taking blobs of text and converting them into database/spreadsheet rows so that they can be used in data science, causal modeling, etc. If you understand Python and can imagine simulating your complex system in Python, then this session is for you! 


Due to the mathematical nature of probabilistic programming systems, participants are expected to have existing knowledge in calculus and be comfortable with basic concepts related to working with probabilities and statistics. Experience with one or more numerical programming languages (such as R, Matlab, Python/Numpy, etc...) is desirable.

Participants who include a brief proposal with their application of a problem that they would like to solve using a probabilistic programming system will be given preference in the participant selection process.


Housing for the two-week period will be provided as close as possible to the government per diem rate for Arlington, Virginia. Participants are expected to cover their own travel and housing costs. A limited amount of financial assistance may be available for currently enrolled University students.

Location: The Westin Arlington Gateway


We are no longer accepting applications to the PPAML 2017 Summer School.

Please send your questions about the summer school and application process to