Greedy function approximation: A gradient boosting machine.

@article{Friedman2001GreedyFA,
  title={Greedy function approximation: A gradient boosting machine.},
  author={J. Friedman},
  journal={Annals of Statistics},
  year={2001},
  volume={29},
  pages={1189-1232}
}
  • J. Friedman
  • Published 2001
  • Mathematics
  • Annals of Statistics
  • 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 and steepest-descent minimization. A general gradient descent boosting paradigm is developed for additive expansions based on any fitting criterion. Specific algorithms are presented for least-squares, least absolute deviation, and Huber-M loss functions for regression, and multiclass logistic… CONTINUE READING
    9,851 Citations

    Figures and Tables from this paper

    Optimization by gradient boosting
    • 9
    • Highly Influenced
    • PDF
    Gradient Boosting Trees
    • PDF
    Stochastic gradient boosting
    • 3,319
    • PDF
    A Fast Sampling Gradient Tree Boosting Framework
    • Highly Influenced
    • PDF
    Conjugate direction boosting for regression
    • 2004
    • 1
    • Highly Influenced
    • PDF
    Robust boosting for regression problems
    • Highly Influenced
    • PDF
    BOOSTING ALGORITHMS: REGULARIZATION, PREDICTION AND MODEL FITTING
    • 741
    • Highly Influenced
    • PDF
    Special Invited Paper-Additive logistic regression: A statistical view of boosting
    • 4,090
    • PDF
    Boosting with early stopping: Convergence and consistency
    • 318
    • PDF
    Gradient and Newton Boosting for Classification and Regression
    • 13
    • Highly Influenced
    • PDF

    References

    SHOWING 1-10 OF 32 REFERENCES
    Additive Logistic Regression : a Statistical
    • 666
    A Geometric Approach to Leveraging Weak Learners
    • 36
    Generalized Additive Models
    • 4,118
    • PDF
    Improved Boosting Algorithms using Confidence-Rated Predictions
    • 2,055
    • Highly Influential
    • PDF
    Experiments with a New Boosting Algorithm
    • 8,004
    • Highly Influential
    • PDF
    Prediction Games and Arcing Algorithms
    • L. Breiman
    • Mathematics, Computer Science
    • Neural Computation
    • 1999
    • 518
    • Highly Influential
    • PDF
    Radial Basis Functions
    • 357
    • PDF
    Radial basis functions
    • 1,006
    Learning representations by back-propagating errors
    • 16,999
    • Highly Influential
    • PDF
    Improving Regressors using Boosting Techniques
    • 469
    • Highly Influential
    • PDF