Fast, Accurate, and Simple Models for Tabular Data via Augmented Distillation
arXiv:2006.14284
Abstract
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.
References in corpus (12)
- Distilling the Knowledge in a Neural Network
- Wide & Deep Learning for Recommender Systems
- AutoGluon-Tabular: Robust and Accurate AutoML for Structured Data
- Understanding and Improving Knowledge Distillation
- Born-Again Tree Ensembles
- Autoregressive Energy Machines
- Learning Discrete Distributions by Dequantization
- Knowledge distillation via adaptive instance normalization
- Convergence rate and concentration inequalities for Gibbs sampling in high dimension
- TraDE: Transformers for Density Estimation
- Learning from weakly dependent data under Dobrushin's condition
- Gradient Boosted Decision Tree Neural Network