We report on a method for compiling decision trees into weighted finite-state transducers. The key assumptions are that the tree predictions specify how to rewrite symbols from an...
We observe that pushdown tree automata (PTAs) known in the literature cannot express combinations of branching and pushdown properties. This is because a PTA processes the children...
In this paper, a new learning framework?probabilistic boosting-tree (PBT), is proposed for learning two-class and multi-class discriminative models. In the learning stage, the pro...