Seminario "Adaptive Decision Tree Methods"
Miércoles 27/11, 17.15h
Presentado por Matias D. Cattaneo
Paper abstract
Adaptive Decision Trees are a popular methodology in modern machine learning and data science. While they are widely used in practice, many of their theoretical and methodological properties remain unknown. This talk presents negative and positive results concerning the statistical properties of different variants of adaptive decision tree procedures. First, it is demonstrated that classical adaptive decision trees implemented using CART are pointwise (and hence uniformly) over the feature space inconsistent. For example, it is discussed how this finding implies important negative implications for heterogeneous causal inference analysis or personalized recommendation systems. Second, it is shown that adaptive oblique trees achieve near optimal mean square convergence under specific conditions, making them competitive relative to one-layer neural network procedures.
The talk will be based on the following work of papers:
Inference with Mondrian Random Forests
Convergence Rates Of Oblique Regression Trees For Flexible Function Libraries
On the Pointwise Behavior of Recursive Partitioning and Its Implications for Heterogeneous Causal Effect Estimation
Matías D. Cattaneo
Ph.D. in Economics from the University of California at Berkeley. Professor of Operations Research and Financial Engineering (ORFE) at Princeton University.
Matías is an elected Fellow of the American Statistical Association, the Institute of Mathematical Statistics, and the International Association for Applied Econometrics. He serves in the editorial boards of the Journal of the American Statistical Association, Econometrica, Operations Research, Statistical Science, the Econometrics Journal, the Journal of Econometrics, Econometric Theory, and the Journal of Causal Inference.
Adaptive Decision Trees are a popular methodology in modern machine learning and data science. While they are widely used in practice, many of their theoretical and methodological properties remain unknown. This talk presents negative and positive results concerning the statistical properties of different variants of adaptive decision tree procedures. First, it is demonstrated that classical adaptive decision trees implemented using CART are pointwise (and hence uniformly) over the feature space inconsistent. For example, it is discussed how this finding implies important negative implications for heterogeneous causal inference analysis or personalized recommendation systems. Second, it is shown that adaptive oblique trees achieve near optimal mean square convergence under specific conditions, making them competitive relative to one-layer neural network procedures.
The talk will be based on the following work of papers:
Inference with Mondrian Random Forests
Convergence Rates Of Oblique Regression Trees For Flexible Function Libraries
On the Pointwise Behavior of Recursive Partitioning and Its Implications for Heterogeneous Causal Effect Estimation
Matías D. Cattaneo
Ph.D. in Economics from the University of California at Berkeley. Professor of Operations Research and Financial Engineering (ORFE) at Princeton University.
Matías is an elected Fellow of the American Statistical Association, the Institute of Mathematical Statistics, and the International Association for Applied Econometrics. He serves in the editorial boards of the Journal of the American Statistical Association, Econometrica, Operations Research, Statistical Science, the Econometrics Journal, the Journal of Econometrics, Econometric Theory, and the Journal of Causal Inference.