Concordance based Survival Cobra with regression type weak learners

TL;DR

This paper introduces a novel survival prediction method combining regression strategies with weak learners, optimizing concordance to improve accuracy in right-censored data scenarios, demonstrated on real datasets.

Contribution

It proposes a new survival prediction approach using concordance maximization with regression-type weak learners and introduces a weighted predictor based on the concordance index.

Findings

Effective survival prediction on real datasets

Improved concordance index scores

Novel weighted predictor formulation

Abstract

In this paper, we predict conditional survival functions through a combined regression strategy. We take weak learners as different random survival trees. We propose to maximize concordance in the right-censored set up to find the optimal parameters. We explore two approaches, a usual survival cobra and a novel weighted predictor based on the concordance index. Our proposed formulations use two different norms, say, Max-norm and Frobenius norm, to find a proximity set of predictions from query points in the test dataset. We illustrate our algorithms through three different real-life dataset implementations.