irboost (version 0.1-1.5)

irb.train_aft: fit a robust accelerated failure time model with iteratively reweighted boosting algorithm

Description

Fit an accelerated failure time model with the iteratively reweighted convex optimization (IRCO) that minimizes the robust loss functions in the CC-family (concave-convex). The convex optimization is conducted by functional descent boosting algorithm in the R package xgboost. The iteratively reweighted boosting (IRBoost) algorithm reduces the weight of the observation that leads to a large loss; it also provides weights to help identify outliers. For time-to-event data, an accelerated failure time model (AFT model) provides an alternative to the commonly used proportional hazards models. Note, function irboost_aft was developed to facilitate a data input format used with function xgb.train for objective=survival:aft in package xgboost. In other ojective functions, the input format is different with function xgboost at the time.

Usage

irb.train_aft(
  params = list(),
  data,
  z_init = NULL,
  cfun = "ccave",
  s = 1,
  delta = 0.1,
  iter = 10,
  nrounds = 100,
  del = 1e-10,
  trace = FALSE,
  ...
)

Value

An object of class xgb.Booster with additional elements:

• weight_update_log a matrix of nobs row by iter column of observation weights in each iteration of the IRCO algorithm

• weight_update a vector of observation weights in the last IRCO iteration that produces the final model fit

• loss_log sum of loss value of the composite function cfun(survival_aft_distribution) in each IRCO iteration

Arguments

params

the list of parameters used in xgb.train of xgboost.
Must include aft_loss_distribution, aft_loss_distribution_scale, but there is no need to include objective. The complete list of parameters is available in the online documentation.

data

training dataset. irboost_aft accepts only an xgb.DMatrix as the input.

z_init

vector of nobs with initial convex component values, must be non-negative with default values = weights if provided, otherwise z_init = vector of 1s

cfun

concave component of CC-family, can be "hacve", "acave", "bcave", "ccave", "dcave", "ecave", "gcave", "hcave".
See Table 2 at https://arxiv.org/pdf/2010.02848.pdf

s

tuning parameter of cfun. s > 0 and can be equal to 0 for cfun="tcave". If s is too close to 0 for cfun="acave", "bcave", "ccave", the calculated weights can become 0 for all observations, thus crash the program

delta

a small positive number provided by user only if cfun="gcave" and 0 < s <1

iter

number of iteration in the IRCO algorithm

nrounds

boosting iterations in xgb.train within each IRCO iteration

del

convergency criteria in the IRCO algorithm, no relation to delta

trace

if TRUE, fitting progress is reported

...

other arguments passing to xgb.train

Author

Zhu Wang
Maintainer: Zhu Wang

References

Wang, Zhu (2021), Unified Robust Boosting, arXiv eprint, https://arxiv.org/abs/2101.07718

See Also

irboost

Examples

# \donttest{
library("xgboost")
X <- matrix(1:5, ncol=1)

# Associate ranged labels with the data matrix.
# This example shows each kind of censored labels.
#          uncensored  right  left  interval
y_lower = c(10,  15, -Inf, 30, 100)
y_upper = c(Inf, Inf,   20, 50, Inf)
dtrain <- xgb.DMatrix(data=X, label_lower_bound=y_lower, label_upper_bound=y_upper)
                      params = list(objective="survival:aft", aft_loss_distribution="normal",
                      aft_loss_distribution_scale=1, max_depth=3, min_child_weight= 0)
watchlist <- list(train = dtrain)
bst <- xgb.train(params, data=dtrain, nrounds=15, watchlist=watchlist)
predict(bst, dtrain)
bst_cc <- irb.train_aft(params, data=dtrain, nrounds=15, watchlist=watchlist, cfun="hcave", 
                       s=1.5, trace=TRUE, verbose=0)
bst_cc$weight_update
predict(bst_cc, dtrain)
# }