It looks like GAMI-Tree is an extension/variant of EBM and GAMI-Net. GAMI-Tree utilizes model-based trees and boosting for describing main and interaction effects instead of neural networks, and there are a few additional algorithms for interaction filtering and iterative fitting. Perhaps it is more computationally efficient, though there is no speed comparison benchmark in the paper. I wonder if there is any plan to implement this method in PiML-Toolbox.
It looks like GAMI-Tree is an extension/variant of EBM and GAMI-Net. GAMI-Tree utilizes model-based trees and boosting for describing main and interaction effects instead of neural networks, and there are a few additional algorithms for interaction filtering and iterative fitting. Perhaps it is more computationally efficient, though there is no speed comparison benchmark in the paper. I wonder if there is any plan to implement this method in PiML-Toolbox.