No Arabic abstract
Gradient boosted decision trees (GBDTs) are widely used in machine learning, and the output of current GBDT implementations is a single variable. When there are multiple outputs, GBDT constructs multiple trees corresponding to the output variables. The correlations between variables are ignored by such a strategy causing redundancy of the learned tree structures. In this paper, we propose a general method to learn GBDT for multiple outputs, called GBDT-MO. Each leaf of GBDT-MO constructs predictions of all variables or a subset of automatically selected variables. This is achieved by considering the summation of objective gains over all output variables. Moreover, we extend histogram approximation into multiple output case to speed up the training process. Various experiments on synthetic and real-world datasets verify that GBDT-MO achieves outstanding performance in terms of both accuracy and training speed. Our codes are available on-line.
In this paper we recreate, and improve, the binary classification method for particles proposed in Roe et al. (2005) paper Boosted decision trees as an alternative to artificial neural networks for particle identification. Such particles are tau neutrinos, which we will refer to as background, and electronic neutrinos: the signal we are interested in. In the original paper the preferred algorithm is a Boosted decision tree. This is due to its low effort tuning and good overall performance at the time. Our choice for implementation is a deep neural network, faster and more promising in performance. We will show how, using modern techniques, we are able to improve on the original result, both in accuracy and in training time.
Precision photometric redshifts will be essential for extracting cosmological parameters from the next generation of wide-area imaging surveys. In this paper we introduce a photometric redshift algorithm, ArborZ, based on the machine-learning technique of Boosted Decision Trees. We study the algorithm using galaxies from the Sloan Digital Sky Survey and from mock catalogs intended to simulate both the SDSS and the upcoming Dark Energy Survey. We show that it improves upon the performance of existing algorithms. Moreover, the method naturally leads to the reconstruction of a full probability density function (PDF) for the photometric redshift of each galaxy, not merely a single best estimate and error, and also provides a photo-z quality figure-of-merit for each galaxy that can be used to reject outliers. We show that the stacked PDFs yield a more accurate reconstruction of the redshift distribution N(z). We discuss limitations of the current algorithm and ideas for future work.
Recurrence data arise from multi-disciplinary domains spanning reliability, cyber security, healthcare, online retailing, etc. This paper investigates an additive-tree-based approach, known as Boost-R (Boosting for Recurrence Data), for recurrent event data with both static and dynamic features. Boost-R constructs an ensemble of gradient boosted additive trees to estimate the cumulative intensity function of the recurrent event process, where a new tree is added to the ensemble by minimizing the regularized L2 distance between the observed and predicted cumulative intensity. Unlike conventional regression trees, a time-dependent function is constructed by Boost-R on each tree leaf. The sum of these functions, from multiple trees, yields the ensemble estimator of the cumulative intensity. The divide-and-conquer nature of tree-based methods is appealing when hidden sub-populations exist within a heterogeneous population. The non-parametric nature of regression trees helps to avoid parametric assumptions on the complex interactions between event processes and features. Critical insights and advantages of Boost-R are investigated through comprehensive numerical examples. Datasets and computer code of Boost-R are made available on GitHub. To our best knowledge, Boost-R is the first gradient boosted additive-tree-based approach for modeling large-scale recurrent event data with both static and dynamic feature information.
Multi-layered representation is believed to be the key ingredient of deep neural networks especially in cognitive tasks like computer vision. While non-differentiable models such as gradient boosting decision trees (GBDTs) are the dominant methods for modeling discrete or tabular data, they are hard to incorporate with such representation learning ability. In this work, we propose the multi-layered GBDT forest (mGBDTs), with an explicit emphasis on exploring the ability to learn hierarchical representations by stacking several layers of regression GBDTs as its building block. The model can be jointly trained by a variant of target propagation across layers, without the need to derive back-propagation nor differentiability. Experiments and visualizations confirmed the effectiveness of the model in terms of performance and representation learning ability.
In this paper, we propose a gradient boosting algorithm for large-scale regression problems called textit{Gradient Boosted Binary Histogram Ensemble} (GBBHE) based on binary histogram partition and ensemble learning. From the theoretical perspective, by assuming the H{o}lder continuity of the target function, we establish the statistical convergence rate of GBBHE in the space $C^{0,alpha}$ and $C^{1,0}$, where a lower bound of the convergence rate for the base learner demonstrates the advantage of boosting. Moreover, in the space $C^{1,0}$, we prove that the number of iterations to achieve the fast convergence rate can be reduced by using ensemble regressor as the base learner, which improves the computational efficiency. In the experiments, compared with other state-of-the-art algorithms such as gradient boosted regression tree (GBRT), Breimans forest, and kernel-based methods, our GBBHE algorithm shows promising performance with less running time on large-scale datasets.