Predictive learning via rule ensembles (2008)
Friedman, Jerome H., Popescu, Bogdan E.
General regression and classification models are constructed as linear combinations of simple rules derived from the data. Each rule consists of a conjunction of a small number of simple statements...
[6] The Chemical Rubber Company. Standard Mathematical Tables, 21st edition, 1973. (2008)
Leo Breiman, Jerome H. Friedman, J. A. Olshen, All S. Collica, Jill P. Card, ...
[1] Nick Atchison and Ron Ross. Wafer zone based yield analysis. In International
Rejoinder: Discussion of Bump Hunting in High Dimensional Data (2007)
Jerome H. Friedman, Nicholas I. Fisher
We thank all of the discussants for taking the time to contribute, and for both their compliments and helpful criticism. Before responding in detail, we address two general concerns that ran...
Comment: Classifier Technology and the Illusion of Progress (2006)
Comment on Classifier Technology and the Illusion of Progress [math.ST/0606441]
Clustering objects on subsets of attributes (2004)
Jerome H. Friedman, Jacqueline J. Meulman
Proofs subject to correction. Not to be reproduced without permission. Confidential until read to the Society. Contributions to the discussion must not exceed 400 words. Contributions longer than 400...
Gradient directed regularization for linear regression and classification (2004)
Jerome H. Friedman, Bogdan E. Popescu
Regularization in linear modeling is viewed as a two–stage process. First a set of candidate models is de…ned by a path through the space of joint parameter values, and then a point on this path...
Gradient Directed Regularization (2004)
Jerome H. Friedman, Bogdan E. Popescu
Regularization in linear regression and classification is viewed as a two-stage process. First a set of candidate models is defined by a path through...
John W. Tukey's work on interactive graphics (2002)
Friedman, Jerome H., Stuetzle, Werner
If there ever was a tool that could stimulate the imagination and profit from the intuition and creativity of John Tukey, it was computer graphics. John always saw graphics a being central to...
Bentley,Jon Louis, Friedman,Jerome H., Maurer,H. A.
This report contains two independent papers on range searching. A range search retrieves from a file all records which conjunctively satisfy a set of range requirements for the keys; that is, each...
A Nested Partitioning Procedure for Numerical Multiple Integration. (2002)
Friedman,Jerome H., Wright,Margaret H.
An algorithm is presented for adaptively partitioning a multidimensional coordinate space based on optimization of a scalar function of the coordinates. The goal is to construct a set of...
Estimating Optimal Transformations for Multiple Regression and Correlation. (2002)
Breiman,Leo, Friedman,Jerome H.
Nonlinear transformation of variables is a commonly used practice in regression problems. Two common goals are stabilization of error variance and asymmetrization/normalization of error distribution....
An Adaptive Importance Sampling Procedure. (2002)
Friedman,Jerome H., Wright,Margaret H.
Monte Carlo calculations often require generation of a random sample of n-dimensional points drawn from a specified multivariate probability distribution. We present an importance sampling technique...
Getting Started with MART (2002)
Multiple additive regression trees (MART) is a methodology for predictive data mining (regression and classification). This note illustrates the use of the R/MART interface. It is intended to be a...
Greedy function approximation: A gradient boosting machine. (2001)
Function estimation/approximation is viewed from the perspective of numerical optimization in function space, rather than parameter space. A connection is made between stagewise additive expansions...
NETWORK DYNAMICS: THEWORLDWIDEWEB (2001)
A. Adamic, Bernardo A. Huberman, Jerome H. Friedman
that I have read this dissertation and that in
Greedy function approximation: A gradient boosting machine (2001)
Function approximation is viewed from the perspective of numerical optimization in function space, rather than parameter space. A connection is made between stagewise additive expansions and...
Stochastic Gradient Boosting (1999)
Gradient boosting constructs additive regression models by sequentially fitting a simple parameterized function (base learner) to current "pseudo"--residuals by least--squares at each...
Greedy Function Approximation: A Gradient Boosting Machine (1999)
Function approximation is viewed from the perspective of numerical optimization in function space, rather than parameter space. A connection is made between stagewise additive expansions and...
Innovations In Local Modeling For Time Series Prediction (1999)
James McNames, Bernard Widrow, Jerome H. Friedman, Jonathan P. How
Previous studies have shown that local models are among the most accurate methods for predicting chaotic time series. This work discusses a number of improvements to local models that reduce...
Tutorial: Getting Started with MART in Splus (1999)
Multiple additive regression trees (MART) is a methodology for predictive data mining (regression and classification). This note illustrates the use of the MART/Splus interface. It is intended to be...
Tutorial: Getting Started with MART in Splus (1999)
Jerome Friedman Stanford, Jerome H. Friedman
Multiple additive regression trees (MART) is a methodology for predictive data mining (regression and classification). This note illustrates the use of the MART/Splus interface. It is intended to be...
Hardware for Kinematic Statistical Graphics. (1998)
Friedman,Jerome H., Stuetzle,Werner
The hardware requirements for a computer graphics system capable of supporting kinematic statistical graphics are specified. The various options are discussed, and the ORION-1 workstation currently...
Multidimensional Additive Spline Approximation. (1998)
Friedman,Jerome H., Stuetzle,Werner, Grosse,Eric
Sponsored in part by Contract DE-AT03-81-ER108-43 and Grant NSF-MCS78-17697.
Projection Pursuit Density Estimation. (1998)
Friedman,Jerome H., Stuetzle,Werner, Schroeder,Anne
The projection pursuit methodology is applied to the multivariate density estimation problem. The resulting nonparametric procedure is often less biased than kernel and near neighbor methods and does...
Smoothing of Scatterplots. (1998)
Friedman,Jerome H., Stuetzle,Werner
A variable span scatterplot smoother based on local linear fits is described. Local cross-validation is used to estimate the optimal span as a function of abscissa value. A rejection rule is...
An Introduction to Real Time Graphical Techniques for Analyzing Multivariate Data. (1998)
Friedman,Jerome H., McDonald,John Alan, Stuetzle,Werner
Orion I is a graphics system used to study applications of computer graphics -- especially interactive motion graphics -- in statistics. Orion I is the newest of a family of 'Prim' systems, whose...
Projection Pursuit Methods for Data Analysis. (1998)
Friedman,Jerome H., Stuetzle,Werner
The report describes new procedures for multivariate regression and density estimation. The procedures construct models for regression surfaces and densities based on the information contained in...
Final Technical Reports 15 June 1983 through 31 March 1986 on Contract N00014-83-K-0472, (1998)
Faculty and graduate students supported by this contract have been instrumental in developing procedures for data analysis. Two such procedures are 'projection pursuit' in which higher dimensional...
On bias, variance, 0/1-loss, and the curse-of-dimensionality (1997)
Jerome H. Friedman, Usama Fayyad
Abstract. The classification problem is considered in which an output variable y assumes discrete values with respective probabilities that depend upon the simultaneous values of a set of input...
Jerome H. Friedman, Ron Kohavi, Yeogirl Yun
Lazy learning algorithms, exemplified by nearestneighbor algorithms, do not induce a concise hypothesis from a given training set; the inductive process is delayed until a test instance is given....
Charles B. Roosen, Jerome H. Friedman, Art B. Owen
In recent years the statistical and engineering communities have developed many high-dimensional methods for regression (e.g. MARS, feedforward neural networks, projection pursuit). Users of these...
Flexible Metric Nearest Neighbor Classification (1994)
The K-nearest-neighbor decision rule assigns an object of unknown class to the plurality class among the K labeled "training" objects that are closest to it. Closeness is usually defined in...
PRODUCTION OF S = 0, -1 RESONANT STATES IN K-p INTERACTIONS AT 2.45 GeV/c (1964)
Ross, Ronald R., Friedman, Jerome H., Siegel, Daniel M., Flatte, Stanley, Alvarez, Luis W., Barbaro-Galtieri, Angela, ...
On bagging and nonlinear estimation (0000)
We propose an elementary model for the way in which stochastic perturbations of a statistical objective function, such as a negative log-likelihood, produce excessive nonlinear variation of the...
Clustering objects on subsets of attributes (with discussion)
Jerome H. Friedman, Jacqueline J. Meulman
A new procedure is proposed for clustering attribute value data. When used in conjunction with conventional distance-based clustering algorithms this procedure encourages those algorithms to detect...
On bagging and nonlinear estimation
We propose an elementary model for the way in which stochastic perturbations of a statistical objective function, such as a negative log-likelihood, produce excessive nonlinear variation of the...