Fast and Flexible Monotonic Functions with Ensembles of Lattices

76 citations recorded by the source corpus.

How researchers see this paper

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

Optimization with Non-Differentiable Constraints with Applications to Fairness, Recall, Churn, and Other Goals (2018, methodology): “in some of the experiments of Canini et al. (2016), the models are trained with around 100,000 constraints, but each constraint only touches two model parameters. Problems like that with many cheap sparse constraints can be well-handled by stochastic sampling of the constraints, as in Cotter et al. ”
How do Humans Understand Explanations from Machine Learning Systems? An Evaluation of the Human-Interpretability of Explanation (2018, background): “, 2012, Chandrashekar and Sahin, 2014], monotone functions [Canini et al., 2016], or efficient logic-based models [Rivest, 1987].”
Proxy Fairness (2018, methodology): “There were N = 5 , 242 , 880 training and 1 , 048 , 576 testing examples, and the underlying classiﬁer is a nonlinear ensemble of 20 x 8 -dimensional lattices [4] trained on D = 20 features.”
An Evaluation of the Human-Interpretability of Explanation (2019, background): “, 2012, Chandrashekar and Sahin, 2014], monotone functions [Canini et al., 2016], or efficient logic-based models [Rivest, 1987].”
Training Well-Generalizing Classifiers for Fairness Metrics and Other Data-Dependent Constraints (2018, methodology): “On each dataset, we trained one of three different types of models: linear, RTL (Canini et al., 2016), or a neural network with one hidden layer containing 50 ReLU neurons.”