Cautious weighted random forests

Zhang, Haifei; Quost, Benjamin; Masson, Marie-Hélène

doi:10.1016/j.eswa.2022.118883

Cited by 14 publications

(10 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SVM is a robust algorithm with mature theory and high efficiency, due to which it has a good application effect in multiple fields 34 , 35 . On the other hand, RF is known for reducing variance error and can be trained in parallel to improve computational efficiency 36 . These robust and multi-dimensional aspects of CART, RF and SVM make them diverse base learners.…”

Section: Proposed Methodologymentioning

confidence: 99%

A multi-objective stacked regression method for distance based colour measuring device

Brar,

Singh

2024

Sci Rep

View full text Add to dashboard Cite

Identifying colour from a distance is challenging due to the external noise associated with the measurement process. The present study focuses on developing a colour measuring system and a novel Multi-target Regression (MTR) model for accurate colour measurement from distance. Herein, a novel MTR method, referred as Multi-Objective Stacked Regression (MOSR) is proposed. The core idea behind MOSR is based on stacking as an ensemble approach with multi-objective evolutionary learning using NSGA-II. A multi-objective optimization approach is used for selecting base learners that maximises prediction accuracy while minimising ensemble complexity, which is further compared with six state-of-the-art methods over the colour dataset. Classification and regression tree (CART), Random Forest (RF) and Support Vector Machine (SVM) were used as regressor algorithms. MOSR outperformed all compared methods with the highest coefficient of determination values for all three targets of the colour dataset. Rigorous comparison with state-of-the-art methods over 18 benchmarked datasets showed MOSR outperformed in 15 datasets when CART was used as a regressor algorithm and 11 datasets when RF and SVM were used as regressor algorithms. The MOSR method was statistically superior to compared methods and can be effectively used to measure accurate colour values in the distance-based colour measuring device.

show abstract

Section: Proposed Methodologymentioning

confidence: 99%

A multi-objective stacked regression method for distance based colour measuring device

Brar,

Singh

2024

Sci Rep

View full text Add to dashboard Cite

show abstract

“…The procedure described in Alg. 3, hereafter referred to as CDM Vote (standing for "cautious decision-making via voting"), extends voting when votes are expressed as subsets of classes and returns the subset A ⋆ = arg max EU (A) among all subsets A ⊆ Ω such that |A| ≤ M ≤ K. It generalizes the method proposed in [24,25] for binary cautious classification. It is computationally less efficient than CDM Ave, even if time complexity can be controlled, as it will be shown in the experimental part.…”

Section: Generalization Of Votingmentioning

confidence: 99%

“…In [24,25], we have proposed a generalized voting aggregation strategy for binary cautious classification within the belief function framework. In the present paper, we generalize these previous works in the multi-class case.…”

Section: Introductionmentioning

confidence: 99%

Cautious Decision-Making for Tree Ensembles

Zhang,

Quost,

Masson

2023

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Cautious classifiers are designed to make indeterminate decisions when the uncertainty on the input data or the model output is too high, so as to reduce the risk of making wrong decisions. In this paper, we propose two cautious decision-making procedures, by aggregating trees providing probability intervals constructed via the imprecise Dirichlet model. The trees are aggregated in the belief functions framework, by maximizing the lower expected discounted utility, so as to achieve a good compromise between model accuracy and determinacy. They can be regarded as generalizations of the two classical aggregation strategies for tree ensembles, i.e., averaging and voting. The efficiency and performance of the proposed procedures are tested on random forests and illustrated on three UCI datasets.

show abstract

“…These bounds are obtained using the Imprecise Dirichlet Model, and reflect the estimation uncertainty due to the lack of training data. These intervals can be pooled using the theory of belief functions, by computing the belief and plausibility 𝑏𝑒𝑙 (𝑌 = 1|𝑥) and 𝑝𝑙 (𝑌 = 1|𝑥), which can then be used in a cautious decision-making process such as interval dominance, possibly resulting in indeterminate decisions (Zhang et al, 2021).…”

Section: Cautious Random Forestsmentioning

confidence: 99%

“…However, when training data are scarce, or when mistakes have a very high cost, cautious classifiers can alternatively be used to provide set-valued decisions rather than single classes and thus control the risk. Cautious random forests (CRF) (Zhang et al, 2021) are one of those classifiers. A CRF combines the classical random forest (RF) strategy (Breiman, 2001), the Imprecise Dirichlet Model (IDM) (Walley, 1996) and the theory of belief functions (Shafer, 1976).…”

Section: Introductionmentioning

confidence: 99%

Explaining Cautious Random Forests via Counterfactuals

Zhang

Quost

Masson

2022

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

Cautious random forests are designed to make indeterminate decisions when tree outputs are conflicting. Since indeterminacy has a cost, it seems desirable to highlight why a precise decision could not be made for an instance, or which minimal modifications can be made to the instance so that the decision becomes a single class. In this paper, we apply an efficient extractor to generate determinate counterfactual examples of different classes, which are used to explain indeterminacy. We evaluate the efficiency of our strategy on different datasets and we illustrate it on two simple case studies involving both tabular and image data.

show abstract

Cautious weighted random forests

Cited by 14 publications

References 39 publications

A multi-objective stacked regression method for distance based colour measuring device

A multi-objective stacked regression method for distance based colour measuring device

Cautious Decision-Making for Tree Ensembles

Explaining Cautious Random Forests via Counterfactuals

Contact Info

Product

Resources

About