Timezone: »

Fast, Accurate, and Simple Models for Tabular Data via Augmented Distillation
Rasool Fakoor · Jonas Mueller · Nick Erickson · Pratik Chaudhari · Alexander Smola

Wed Dec 09 09:00 PM -- 11:00 PM (PST) @ Poster Session 4 #1319

Automated machine learning (AutoML) can produce complex model ensembles by stacking, bagging, and boosting many individual models like trees, deep networks, and nearest neighbor estimators. While highly accurate, the resulting predictors are large, slow, and opaque as compared to their constituents. To improve the deployment of AutoML on tabular data, we propose FAST-DAD to distill arbitrarily-complex ensemble predictors into individual models like boosted trees, random forests, and deep networks. At the heart of our approach is a data augmentation strategy based on Gibbs sampling from a self-attention pseudolikelihood estimator. Across 30 datasets spanning regression and binary/multiclass classification tasks, FAST-DAD distillation produces significantly better individual models than one obtains through standard training on the original data. Our individual distilled models are over 10x faster and more accurate than ensemble predictors produced by AutoML tools like H2O/AutoSklearn.

Author Information

Rasool Fakoor (Amazon AWS)
Jonas Mueller (Amazon Web Services)
Nick Erickson (Amazon Web Services)
Pratik Chaudhari (University of Pennsylvania)
Alex Smola (Amazon - We are hiring!)

**Amazon AWS Machine Learning** We are hiring!

More from the Same Authors