這是一篇來自于 微軟和卡耐基的文章复旬，iccv2015best paper 以及其對應(yīng)的 材料supplemental materials
主要貢獻(xiàn)
- A novel approach that unifies classification trees with the representation learning functionality known from deep convolution networks, by training them in an end-to-end manner.
- Provide a globally optimal strategy to estimate predictions taken in the leaves.
決策樹示意圖
決策樹示意圖
