Generalized Boosted Models: A guide to the gbm package

By Greg Ridgeway. 2006

948 citations recorded by the source corpus.

Subfields: Statistics and Probability, Optimization & Training

How researchers see this paper

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

In Statistics and Probability

In Optimization & Training

Citations in the corpus

Interpretable classification models for recidivism prediction (2015, methodology): “et al., 2012) 19 values of W+ C2(0:01;0:1;1;10) Stochastic Gradient Boosting (Adaboost) Boosting gbm (Ridgeway, 2006) 19 values of W+ shrinkage 2(0:001;0:01;0:1) ntrees 2(100;500;1500;3000) (based on Ridgeway 2007) SLIM Scoring Systems SLIM CPLEX 12.6 19 values of W+; C 0;set to ﬁnd most accurate mo”
XGBoost: A Scalable Tree Boosting System (2016, background): “In order to do so eﬃciently, the algorithm must ﬁrst sort the data according to feature values and visit the data in sorted order to accumulate the gradient statistics for the structure score in Eq (7).”
Propensity score estimation: machine learning and classification methods as alternatives to logistic regression (2010, methodology): “Last, implementations of boosting are available for R [39,40], SAS [18], and Stata [41] (Table 1).”
CS2P: Improving Video Bitrate Selection and Adaptation with Data-Driven Throughput Prediction (2016, methodology): “Machine-learning predictors: SVR (Support Vector Regression [34]) and GBR (Gradient Boosting Regression trees [41]).”
Enhancer-promoter interactions are encoded by complex genomic signatures on looping chromatin (2016, methodology): “Alternate software implementation To confirm robustness to underlying software, an alternative version of the TargetFinder pipeline was implemented in R using the caret [2], randomForest [3], gbm [4], and glmnet [5] packages.”
Yahoo! Learning to Rank Challenge Overview (2010, methodology): “Publicly available packages exist (Ridgeway, 2007), but we used our internal implementation (Ye et al., 2009).”
Machine Learning Risk Assessments in Criminal Justice Settings (2018, background): “that perform admirably are gbm (Ridgeway 2007) and XGBoost (Chen and Guestrin 2016).”
A Bias Correction Algorithm for the Gini Variable Importance Measure in Classification Trees (2008, methodology): “2006), the gbm package in R (Ridgeway 2007), the boost Stata command (Schonlau 2005), and the MART package in S-Plus and R (Friedman 2002).”
Estimating forest soil bulk density using boosted regression modelling (2010, methodology): “The optimal number of trees (best.iter) is given by the gbm algorithm and depends on the learning rate (Ridgeway, 2007).”
ToxicBlend: Virtual Screening of Toxic Compounds with Ensemble Predictors (2018, methodology): “We built our blended predictor using multi-task gradient boosting with an additional constraint on the monotonous dependency between the individual model predictions and the ﬁnal combined score [45].”