Event Phone: 1-610-715-0115

Cancellation Policy: If you cancel your registration at least two weeks before the course is scheduled to begin, you are entitled to a full refund (minus a processing fee of $50).

In the unlikely event that Statistical Horizons LLC must cancel a seminar, we will do our best to inform you as soon as possible of the cancellation. You would then have the option of receiving a full refund of the seminar fee or a credit towards another seminar. In no event shall Statistical Horizons LLC be liable for any incidental or consequential damages that you may incur because of the cancellation.

A 4-Day Livestream Seminar Taught by Jens Hainmueller, Ph.D.

A Practical Guide to Unpacking the Black Box

Machine learning often outperforms traditional models like linear or logistic regression in predictive accuracy, but this advantage often comes at the cost of higher complexity and output that can be difficult to explain and interpret.

Accurate prediction is often not enough—researchers need to understand why a model performs well, which features are driving its decisions, and how predictions differ across subgroups. This transparency helps researchers develop fair, reliable, and robust models, and to translate their work to real-world scenarios where explainability is often a requirement.

This course is designed to teach you how to make machine learning models more transparent and interpretable. After reviewing interpretable models such as linear and logistic regression, we will examine several popular machine learning models and demonstrate how they can be made more interpretable using a range of post-hoc and model-agnostic methods that provide insights at both the aggregate and individual levels. These methods include partial dependence plots, Accumulated Local Effects (ALE) plots, feature interaction measures (H-statistic), functional decomposition, permutation feature importance, global surrogate models, individual conditional expectation (ICE) curves, local surrogate models (such as LIME), scoped rules (anchors), counterfactual explanations, Shapley values, and SHAP values.

Throughout the course, core technical concepts will be demonstrated with real-world datasets and hands-on coding exercises. This will ensure that you not only understand the theory behind interpretability but also acquire practical skills to apply these techniques in your own projects.

By the end of the course, you will be equipped with the knowledge and tools necessary to interpret machine learning models effectively, allowing for better insights, improved model transparency, and greater trust in their systems.

Interpretable Machine Learning refers to techniques and approaches that make machine learning models understandable to humans, ensuring that their behavior and predictions are transparent.

We will start by reviewing intrinsically interpretable models, such as linear and logistic regression. These models offer clear interpretations by providing coefficients that explain the importance of each feature, helping us understand how predictions are made. This will set the foundation for the level of interpretability we aim to achieve in more complex models. We will also explore techniques like Lasso regression, which enhance interpretability by enforcing sparsity in the coefficients.

Next, we’ll examine more complex machine learning models that are not inherently interpretable, such as gradient-boosted trees, BART (Bayesian Additive Regression Trees), or neural networks. We will then introduce a range of post-hoc and model-agnostic interpretation methods that help researchers interpret these models. These include both global methods—which explain how features impact predictions on average—and local methods—which focus on individual predictions.

For global interpretation, we will focus on partial dependence plots, which show how a feature influences the prediction when other features are averaged out. These plots are useful for understanding the overall behavior of a model. Additionally, we’ll cover other techniques like Accumulated Local Effects (ALE) plots, feature interaction measures (H-statistic), functional decomposition, permutation feature importance, and global surrogate models.

For local interpretability, we will delve into individual conditional expectation (ICE) curves, which illustrate how changes in a feature affect the prediction for an individual instance. We’ll also explore local surrogate models (such as LIME), which approximate a complex model with a simpler, interpretable one for specific predictions. Further, we will cover scoped rules (anchors), counterfactual explanations, which reveal how features need to change to alter a prediction, and Shapley values, which fairly attribute a prediction to individual features. We’ll also discuss the widely-used SHAP values, which provide consistent and theoretically sound feature attributions.