2018

CatBoost: gradient boosting with categorical features support

Dorogush, Anna Veronika, Ershov, Vasily, Gulin, Andrey

Understand

In this paper we present CatBoost, a new open-sourced gradient boosting library that successfully handles categorical features and outperforms existing publicly available implementations of gradient boosting in terms of quality on a set of popular publicly available datasets.

  • The library has a GPU implementation of learning algorithm and a CPU implementation of scoring algorithm, which are significantly faster than other gradient boosting libraries on ensembles of similar sizes.

Built on

  • Estimating probabilities: a crucial task in machine learning

    B. Cestnik et al · 1990

    Earlier work this paper cites.

  • Oblivious decision trees and abstract cases

    P. Langley and S. Sage · 1994

    Earlier work this paper cites.

  • Oblivious decision trees, graphs, and top-down pruning

    R. Kohavi and C.-H. Li · 1995

    Earlier work this paper cites.

  • Out-of-bag estimation, 1996

    L. Breiman · 1996

    Earlier work this paper cites.

  • Greedy function approximation: a gradient boosting machine

    J. H. Friedman · 2001

    Earlier work this paper cites.

  • A preprocessing scheme for high-cardinality categorical attributes in classification and prediction problems

    D. Micci-Barreca · 2001

    Earlier work this paper cites.

Similar

  • Stochastic gradient boosting

    J. H. Friedman · 2002

    Cited alongside, same era.

  • Boosted decision trees as an alternative to artificial neural networks for particle identification

    B. P. Roe, H.-J. Yang, J. Zhu, Y. Liu, I. Stancu, and G. McGregor · 2005

    Cited alongside, same era.

  • An empirical comparison of supervised learning algorithms

    R. Caruana and A. Niculescu-Mizil · 2006

    Cited alongside, same era.

  • Efficient parallel scan algorithms for gpus

    M. G. Shubhabrata Sengupta, Mark Harris · 2008

    Cited alongside, same era.

  • Adapting boosting for information retrieval measures

    Q. Wu, C. J. Burges, K. M. Svore, and J. Gao · 2010

    Cited alongside, same era.

  • Winning the transfer learning track of yahoo!’s learning to rank challenge with yetirank

    A. Gulin, I. Kuralenok, and D. Pavlov · 2011

    Cited alongside, same era.

Then

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…