Presented By O’Reilly and Cloudera
Make Data Work
March 5–6, 2018: Training
March 6–8, 2018: Tutorials & Conference
San Jose, CA

Human in the loop: Bayesian rules enabling explainable AI

Pramit Choudhary (
2:40pm3:20pm Thursday, March 8, 2018
Average rating: *****
(5.00, 3 ratings)

Who is this presentation for?

  • Data scientists, data analysts, and machine learning practitioners

Prerequisite knowledge

  • A basic understanding of generative and discriminative models, machine learning, and statistical modeling

What you'll learn

  • Understand the concept of model evaluation and interpretability as it relates to enterprise data challenges as well as Bayesian inference
  • Learn best practices for understanding a model's behavior when building predictive modeling pipelines


The adoption of machine learning to solve real-world problems has increased exponentially, but users still struggle to derive full potential of the predictive models. It is no longer sufficient to evaluate a model’s accurate prediction just on a validation set based on error metrics. However, there is still a dichotomy between explainability and model performance when choosing an algorithm. Linear models and simple decision trees are often preferred over more complex models such as ensembles or deep learning models for ease of interpretation, but this often results in loss in accuracy. However, is it actually necessary to accept a trade-off between model complexity and interpretability?

Pramit Choudhary explores the usefulness of a generative approach that applies Bayesian inference to generate human-interpretable decision sets in the form of “if. . .and else” statements. These human interpretable decision lists with high posterior probabilities might be the right way to balance between model interpretability, performance, and computation. This is an extension of’s ongoing effort to enable trust in predictive algorithms to drive better collaboration and communication among peers. Pramit also outlines’s open source model interpretation framework, Skater, and explains how it helps practitioners understand model behavior better without compromising on the choice of algorithm.

Photo of Pramit Choudhary

Pramit Choudhary

Pramit Choudhary is a Lead data scientist/ML scientist at, where he focuses on optimizing and applying classical machine learning and Bayesian design strategy to solve large scale real-world problems.
Currently, he is leading initiatives on figuring out better ways to generate a predictive model’s learned decision policies as meaningful insights(Supervised/Unsupervised problems)

Comments on this page are now closed.


Picture of Pramit Choudhary
Pramit Choudhary | LEAD DATA SCIENTIST
03/10/2018 2:56am PST

Thanks for joining the talk everyone. Feel free to reach out if you have questions or suggestions.
Check out Skater
Will post the slides from the presentation soon