Fast Flux Discriminant (FFD)

FFD is an interpretable machine learning approach to large-scale nonlinear classification. The main idea is to map the data to a new feature space based on kernel smoothing. A linear discriminative model is then learned to optimize the feature weights. For further information, please see our KDD'14 paper [1]. It has the potential of achieving the followings:

Please also note that FFD may not be good at high dimensional data such as text or image. First, high dimensional data doesn't really need nonlinearity. Second, those data might have higher-order interaction between features and it is hard for the FFD model to capture all of them.

Matlab Code Download

For any questions or bug reports, please contact Wenlin Chen (wenlinchen AT


[1]. W. Chen, Y. Chen, and K. Weinberger, Fast Flux Discriminant for Large-Scale Sparse Nonlinear Classification, Proc. ACM SIGKDD Conference (KDD-14), 2014. (paper)


This research is supported in part by the CNS-1017701, CCF-1215302, and IIS-1343896 grants from the National Science Foundation of the United States, a Microsoft Research New Faculty Fellowship, a Washington University URSA grant, and a Barnes-Jewish Hospital Foundation grant, NSF grants 1149882, 1137211.