Multivariate adaptive regression splines

By Jerome H. Friedman. 1990

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

Local Linear Forests (2018, methodology): “2 Related Work Random forests were first introduced by Breiman [2001], building on the work of Breiman, Friedman, Stone, and Olshen [1984] on recursive partitioning (CART), Breiman [1996] on bagging, and Amit and Geman [1997] on randomized trees.”
BART: Bayesian Additive Regression Trees (2008, methodology): “For the predictive modeling problem of TF-DNA binding in genetics, Liu & Zhou (2007) and Zhou & Liu (2007) considered a variety of learning methods, including stepwise linear regression, MARS, neural networks, support vector machines, boosting and BART.”
Hierarchical Mixtures of Experts and the EM Algorithm (1993, methodology): “The CART algorithm [l], the MARS algorithm [ 5 ], and the ID3 algorithm [12] are well-known examples.”
Bayesian regression tree ensembles that adapt to smoothness and sparsity (2017, methodology): “A standard test case, initially proposed by Friedman (1991) (see also Chipman et al., 2010), sets This f 0 ( x ) features two nonlinear terms, two linear terms, with a nonlinear interaction.”
Sample selection bias and presence-only distribution models: implications for background and pseudo-absence data. (2009, background): “Because the regression-related methods and other newer initiatives show generally higher predictive performance than other approaches (e.g., Elith et al. 2006, Hernandez et al. 2006), we focus here on a subset of more successful, widely used methods: boosted regression trees (BRT; Leathwick et al. 2”
Validation requirements for diffuse reflectance soil characterization models with a case study of VNIR soil C prediction in Montana (2005, methodology): “data-mining techniques like Multivariate Adaptive Regression Splines (MARS) (Friedman, 1991) have also been employed to identify and model correlations between correlated predictors and soil properties of interest (Shepherd and Walsh, 2002).”
Free‐knot polynomial splines with confidence intervals (2003, methodology): “Some references which used this approach are Friedman and Silverman (1989), Friedman (1991) and Stone et al. (1997).”
Incremental learning for ν-Support Vector Regression (2015, unknown): “Friedman is an artificial dataset (Friedman, 1991).”
MALTS: Matching After Learning to Stretch (2018, methodology): “We use two main data-generative processes: The first DGP has a linear baseline with linear and quadratic treatment effects while the second DGP is the extension of Friedman’s function introduced to test performance of prediction algorithms of Friedman (1991). This second DGP, also termed as Friedman”
Strong and moderate nonlinear El Niño regimes (2016, methodology): “For this purpose, we use the piecewise linear (hereafter “nonlinear”) regression method based on the Multivariate Adaptive Regression Splines algorithm (Friedman 1991), as implemented by Jekabsons”
A comparison of deep networks with ReLU activation function and linear spline-type methods (2018, methodology): “In the original article Friedman (1991) the values for tj are restricted to a smaller set. The least squares functional can be replaced by any other loss. The runtime of the algorithm is of the order (M ′)3nd, where the key step is the computation of the argmin which is easily parallelizable. Friedm”