Trading Interpretability for Accuracy: Oblique Treed Sparse Additive Models

By Jialei Wang, Ryohei Fujimaki, Yosuke Motohashi. 2015

49 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Safety, Robustness & Interpretability

Citations in the corpus

Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI) (2018, background): “[164] proposed in their work ’’Trading Interpretability for Accuracy’’ an Oblique Treed Sparse Additive Models that sacrifices a certain degree of interpretability for accuracy in order to achieve entirely sufficient accuracy.”
Making Tree Ensembles Interpretable (2016, neutral): “Wang et al. (2015) studied a similar approach with a richly-structured tree model.”
Scalable Bayesian Rule Lists (2016, background): “One could go beyond logical models and consider also mixed logical/linear models (see Wang, Fujimaki, & Motohashi, 2015a).”
Optimal randomized classification trees (2021, background): “A different perspective dealing with oblique splits can be found in [41], where the authors propose the so-called oblique tree sparse additive models.”
Hybrid Decision Making: When Interpretable Models Collaborate With Black-Box Models (2018, background): “A recent work [33] divides feature spaces into regions with sparse oblique tree spliing and assign local sparse additive experts to individual regions.”
Model-Agnostic Linear Competitors - When Interpretable Models Compete and Collaborate with Black-Box Models (2019, background): “A recent work (Wang et al. 2015) divides feature spaces into regions with sparse oblique tree splitting and assign local sparse additive experts to individual regions.”
Variable Selection in Heterogeneous Datasets: A Truncated-rank Sparse Linear Mixed Model with Applications to Genome-wide Association Studies (2017, background): “In applications, variable selection is broadly used to extract variables that are interpretable or potentially causal [9], [10], especially in biology [11] and medicine [12].”
Interpretable Companions for Black-Box Models (2020, methodology): “[7] proposed to divide feature spaces into regions with sparse oblique trees and then assign black-box local experts to each region.”
A Survey on the Explainability of Supervised Machine Learning (2020, methodology): “…Classification Rule-based Ustun and Rudin (2017) Two-level Boolean Rules Classification Rule-based Su et al. (2015) BOA Classification Rule-based Wang et al. (2015b) ORC Classification Rule-based Bertsimas et al. (2011) BLM Classification Rule-based Letham et al. (2012) BRL Classification…”
Recent advances in decision trees: an updated survey (2022, methodology): “The authors reported that this approach was more effective than an earlier model tree algorithm called ELM-trees (Wang et al. 2015b), which instead employed ELM models during pre-pruning and additionally tackled efficiency by employing parallelism in the evaluation of candidate splits.”
Explainable Artificial Intelligence: a Systematic Review (2020, background): “Oblique Treed Sparse Additive Models (OT-SpAMs) [312] are instances of regionspecific predictive models.”