Integrated Brier Score based Survival Cobra -- A regression based
approach
- URL: http://arxiv.org/abs/2210.12006v1
- Date: Fri, 21 Oct 2022 14:48:10 GMT
- Title: Integrated Brier Score based Survival Cobra -- A regression based
approach
- Authors: Rahul Goswami and Arabin Kumar Dey
- Abstract summary: We provide two novel regression-based integrations of combined regression strategy (COBRA) ensemble using Integrated Brier Score to predict conditional survival function.
Our proposition includes a weighted version of all predictions based on Integrated Brier Score score made by all weak learners to predict the final survival function apart from the straight implementation.
- Score: 0.0
- License: http://creativecommons.org/licenses/by-sa/4.0/
- Abstract: In this paper, we provide two novel regression-based integrations of combined
regression strategy (COBRA) ensemble using Integrated Brier Score to predict
conditional survival function. Our proposition includes a weighted version of
all predictions based on Integrated Brier Score score made by all weak learners
to predict the final survival function apart from the straight implementation.
Two different norms (Frobenius and Sup norm) used to figure out the proximity
points in the algorithm. Our implementations consider right-censored data too.
We illustrate the proposed algorithms through few real-life data analysis.
Related papers
- Variational Deep Survival Machines: Survival Regression with Censored Outcomes [11.82370259688716]
Survival regression aims to predict the time when an event of interest will take place, typically a death or a failure.
We present a novel method to predict the survival time by better clustering the survival data and combine primitive distributions.
arXiv Detail & Related papers (2024-04-24T02:16:00Z) - Area-norm COBRA on Conditional Survival Prediction [0.0]
The paper explores a different variation of combined regression strategy to calculate the conditional survival function.
We use regression based weak learners to create the proposed ensemble technique.
The proposed model shows a construction which ensures that it performs better than the Random Survival Forest.
arXiv Detail & Related papers (2023-09-01T12:20:30Z) - Regression with Label Differential Privacy [64.21020761920322]
We derive a label DP randomization mechanism that is optimal under a given regression loss function.
We prove that the optimal mechanism takes the form of a "randomized response on bins"
arXiv Detail & Related papers (2022-12-12T17:41:32Z) - Okapi: Generalising Better by Making Statistical Matches Match [7.392460712829188]
Okapi is a simple, efficient, and general method for robust semi-supervised learning based on online statistical matching.
Our method uses a nearest-neighbours-based matching procedure to generate cross-domain views for a consistency loss.
We show that it is in fact possible to leverage additional unlabelled data to improve upon empirical risk minimisation.
arXiv Detail & Related papers (2022-11-07T12:41:17Z) - Concordance based Survival Cobra with regression type weak learners [0.0]
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.
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.
arXiv Detail & Related papers (2022-09-24T04:10:17Z) - Deep Probabilistic Graph Matching [72.6690550634166]
We propose a deep learning-based graph matching framework that works for the original QAP without compromising on the matching constraints.
The proposed method is evaluated on three popularly tested benchmarks (Pascal VOC, Willow Object and SPair-71k) and it outperforms all previous state-of-the-arts on all benchmarks.
arXiv Detail & Related papers (2022-01-05T13:37:27Z) - Bias-Robust Bayesian Optimization via Dueling Bandit [57.82422045437126]
We consider Bayesian optimization in settings where observations can be adversarially biased.
We propose a novel approach for dueling bandits based on information-directed sampling (IDS)
Thereby, we obtain the first efficient kernelized algorithm for dueling bandits that comes with cumulative regret guarantees.
arXiv Detail & Related papers (2021-05-25T10:08:41Z) - Estimating leverage scores via rank revealing methods and randomization [50.591267188664666]
We study algorithms for estimating the statistical leverage scores of rectangular dense or sparse matrices of arbitrary rank.
Our approach is based on combining rank revealing methods with compositions of dense and sparse randomized dimensionality reduction transforms.
arXiv Detail & Related papers (2021-05-23T19:21:55Z) - Covariate Distribution Aware Meta-learning [3.494950334697974]
We propose a computationally feasible meta-learning algorithm by introducing meaningful relaxations.
We demonstrate the gains of our algorithm over bootstrapped based meta-learning baselines on popular classification benchmarks.
arXiv Detail & Related papers (2020-07-06T05:00:13Z) - Learning Minimax Estimators via Online Learning [55.92459567732491]
We consider the problem of designing minimax estimators for estimating parameters of a probability distribution.
We construct an algorithm for finding a mixed-case Nash equilibrium.
arXiv Detail & Related papers (2020-06-19T22:49:42Z) - Optimal Feature Manipulation Attacks Against Linear Regression [64.54500628124511]
In this paper, we investigate how to manipulate the coefficients obtained via linear regression by adding carefully designed poisoning data points to the dataset or modify the original data points.
Given the energy budget, we first provide the closed-form solution of the optimal poisoning data point when our target is modifying one designated regression coefficient.
We then extend the analysis to the more challenging scenario where the attacker aims to change one particular regression coefficient while making others to be changed as small as possible.
arXiv Detail & Related papers (2020-02-29T04:26:59Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.