Human-Centered Machine Learning

Our April meetup features a presentation, Human-Centered Machine Learning, by Patrick Hall of H2O.ai. Doors open at 6 pm. After a half-hour of networking and refreshments courtesy of meetup sponsor Allegis Group, our program starts at 6:30 pm.

Patrick's presentation illustrates how to combine innovations from several sub-disciplines of machine learning research to train understandable, observationally fair, trustable, and accurate predictive modeling systems. Techniques from research into fair models, directly interpretable Bayesian or constrained machine learning models, and post-hoc explanations can be used to train transparent, observationally fair, and accurate models. Additional techniques from fairness research can be used to check for disparate impact in model predictions and to preprocess data and post-process predictions to ensure the demographic parity of predictive models. Finally, new testing and debugging techniques can increase the trustworthiness of model predictions on unseen data. These techniques create a new and truly human-centered type of machine learning suitable for use in business- and life-critical decision support.

Patrick Hall is senior director for data science products at H2O.ai where he focuses mainly on model interpretability. Patrick is also an adjunct professor in the Department of Decision Sciences at George Washington University, where he teaches graduate classes in data mining and machine learning.