Globally Optimal Fuzzy Decision Trees for Classification and Regression

By Alberto Suárez, James F. Lutsko. 1999

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Neural Prototype Trees for Interpretable Fine-grained Image Recognition (2020, background): “Soft Decision Trees (SDTs) have shown to be more accurate than traditional hard decision trees [23, 24, 46].”
Decision Forests, Convolutional Networks and the Models in-Between (2016, unknown): “[11, 19, 25]).”
Adaptive Neural Trees (2018, methodology): “The very ﬁrst SDT introduced in (Su´arez & Lutsko, 1999) is a speciﬁc case where in our terminology the routers are axis-aligned features, the transformers are identity functions, and the routers are static distributions over classes or linear functions.”
A complete fuzzy decision tree technique (2003, methodology): “However, there are global optimizations based on genetic algorithms [43], similarity measures [53], or backpropagation of neural networks [15,27,44,47].”
End-to-End Learning of Decision Trees and Forests (2019, background): “While Suárez and Lutsko (1999) focused on a fuzzy approach of decision tree, Jordan (1994) introduced hierarchical mixtures of experts.”
Nonparametric direct mapping of rainfall‐runoff relationships: An alternative approach to data analysis and modeling? (2004, methodology): “Partitioning has already been advocated for dealing with the interference issue and poor prior knowledge of a ‘‘global’’ error structure (as, for example, in the differentiation of hydrograph parts used by Wagener et al. [2003] and Freer et al.”
Deep Neural Decision Forests (2015, methodology): “In [35], fuzzy decision trees were presented, including a training mechanism similar to back-propagation in neural networks.”
Hybrid decision tree (2002, methodology): “Suarez and Lutsko (1999) proposed a fuzzy decision tree by superimposing fuzzy structure over the skeleton of CART (Breiman et al., 1984), which was then trained by a procedure similar to backpropagation to globally optimize the fuzzy splits.”
Recent advances in decision trees: an updated survey (2022, methodology): “Suarez and Lutsko (1999) stated that this procedure is analogous to fixing the architecture and then finding the weights in an ANN.”