Corpus ID: 219559362

Survival regression with accelerated failure time model in XGBoost

@article{Barnwal2020SurvivalRW,
  title={Survival regression with accelerated failure time model in XGBoost},
  author={Avinash Barnwal and Hyunsu Cho and Toby Hocking},
  journal={ArXiv},
  year={2020},
  volume={abs/2006.04920}
}
Survival regression is used to estimate the relation between time-to-event and feature variables, and is important in application domains such as medicine, marketing, risk management and sales management. Nonlinear tree based machine learning algorithms as implemented in libraries such as XGBoost, scikit-learn, LightGBM, and CatBoost are often more accurate in practice than linear models. However, existing implementations of tree-based models have offered limited support for survival regression… Expand
Optimizing ROC Curves with a Sort-Based Surrogate Loss Function for Binary Classification and Changepoint Detection
TLDR
This work proposes a convex relaxation of this objective that results in a new surrogate loss function called the AUM, short for Area Under Min(FP, FN), which requires a sort and a sum over the sequence of points on the ROC curve and is efficiently computed and used in a gradient descent learning algorithm. Expand
Analysis of overweight vehicles on asphalt pavement performance using accelerated failure time models
  • Miaomiao Zhang, Hongren Gong, Yuetan Ma, Rui Xiao, Baoshan Huang
  • Environmental Science
  • International Journal of Pavement Engineering
  • 2021
Vehicle overweighting has been one of the key driving forces of deteriorated pavements. The overweighting characteristics that dominate the pavement performance deterioration have been inadequately...
Computing the Hazard Ratios Associated with Explanatory Variables Using Machine Learning Models of Survival Data
TLDR
A novel way to compute HRs from tree-based ML models using the SHapley Additive exPlanation values, which is a locally accurate and consistent methodology to quantify explanatory variables' contribution to predictions. Expand

References

SHOWING 1-10 OF 51 REFERENCES
Fitting Accelerated Failure Time Models in Routine Survival Analysis with R Package aftgee
TLDR
An R package aftgee is described that implements recently developed inference procedures for AFT models with both the rank-based approach and the least squares approach, and uses an induced smoothing procedure that leads to much more efficient computation than the linear programming method. Expand
Flexible boosting of accelerated failure time models
TLDR
A new boosting algorithm for censored time-to-event data which is suitable for fitting parametric accelerated failure time models and closely approximates the estimates obtained from the maximum likelihood method is introduced. Expand
Allowing for mandatory covariates in boosting estimation of sparse high-dimensional survival models
TLDR
A new boosting algorithm for censored time-to-event data is introduced that shares the favorable properties of existing approaches, but uses an offset-based update mechanism that allows for tailored penalization of the covariates under consideration. Expand
Random survival forests
TLDR
This article introduces random survival forests, a random forests method for the analysis of right-censored survival data, and extends Breiman’s random forests (RF) method, showing it to be highly accurate and comparable to state-of-the-art methods. Expand
Bagging Survival Trees
TLDR
A new method to aggregate survival trees in order to obtain better predictions for breast cancer and lymphoma patients is suggested and the aggregated Kaplan-Meier curve of a new observation is defined by theKapler curve of all observations identified by the B leaves containing the new observation. Expand
Survival trees for interval-censored survival data.
TLDR
A survival tree method for interval-censored data based on the conditional inference framework is proposed and it is found that the tree is effective in uncovering underlying tree structure, performs similarly to an interval- censored Cox proportional hazards model fit when thetrue relationship is linear, and performs at least as well as (and in the presence of right-censoring outperforms) the Cox model when the true relationship is not linear. Expand
The accelerated failure time model: a useful alternative to the Cox regression model in survival analysis.
  • L. Wei
  • Mathematics, Medicine
  • Statistics in medicine
  • 1992
TLDR
In this article, some newly developed linear regression methods for analysing failure time observations have sound theoretical justification and can be implemented with an efficient numerical method. Expand
The comparison of proportional hazards and accelerated failure time models in analyzing the first birth interval survival data
Survival analysis is a branch of statistics, which is focussed on the analysis of time- to-event data. In multivariate survival analysis, the proportional hazards (PH) is the most popular model inExpand
Maximum Margin Interval Trees
TLDR
A dynamic programming algorithm is proposed to learn a tree by minimizing a margin-based discriminative objective function, and this algorithm achieves state-of-the-art speed and prediction accuracy in a benchmark of several data sets. Expand
Time-to-Event Prediction with Neural Networks and Cox Regression
TLDR
New methods for time-to-event prediction are proposed by extending the Cox proportional hazards model with neural networks, and is found to be highly competitive, typically yielding the best performance in terms of Brier score and binomial log-likelihood. Expand
...
1
2
3
4
5
...