DeepRED - Rule Extraction from Deep Neural Networks

By Jan Ruben Zilke, Eneldo Loza Mencía, Frederik Janssen. 2016

236 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

Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AI (2019, positive): “One of these few works is DeepRED algorithm [257] , which extends the de-compositional approach to rule extraction (splitting at neuron level) presented in [259] for multi-layer neural network by adding more decision trees and rules.”
Explaining Explanations: An Overview of Interpretability of Machine Learning (2018, methodology): “Other approaches aim to reverse engineer the neural network, notably, the RxREN algorithm, which is used in DeepRED[21].”
Explanation Methods in Deep Learning: Users, Values, Concerns and Challenges (2018, methodology): “This term appears in (Arbatli and Akin, 1997; Markowska-Kaczmar and Wnuk-Lipi´nski, 2004; Zilke et al., 2016; Ribeiro et al., 2016a,b; Andrews et al., 1995; Lakkaraju et al., 2017).”
GraphLIME: Local Interpretable Model Explanations for Graph Neural Networks (2020, methodology): “DeepRED [12] was proposed to extract rules from deep neural networks and generate explanations for predictions, and ANN-DT [13] was proposed to extract binary decision trees from neural networks, it is an interpretable model to interpret the main model.”
Recent Advancements in End-to-End Autonomous Driving Using Deep Learning: A Survey (2023, background): “Recent studies have explored translating deep learning models into decision trees [136], rule-based models [137], or causal models [138].”
Explaining Explanations to Society (2019, background): “These include salience mapping [29, 38], decision trees [39], automatic rule-extraction [1], and inﬂuence functions [16].”
A comprehensive taxonomy for explainable artificial intelligence: a systematic survey of surveys on methods and concepts (2021, methodology): “The rule extraction technique DeepRED [152] (DeepDeepRED [152] Rule Extraction with Decision tree induction) is an example of an eclectic method, so neither fully model-agnostic nor totally reliant on access to model internals.”
Analysis of Explainers of Black Box Deep Neural Networks for Computer Vision: A Survey (2019, methodology): “Similar to DeepRED [69] the rule generation “KnowledgeTron" ( KT ) [70] applied an if-then-rule for each neuron, layer by layer.”
A Survey on the Explainability of Supervised Machine Learning (2020, methodology): “This is the basis for the final rule extraction step (Biswas et al., 2017) (Zilke et al., 2016).”
Principles and Practice of Explainable Machine Learning (2020, methodology): “DeepRED [91] is one of the most popular such techniques, extending CRED.”
How Case-Based Reasoning Explains Neural Networks: A Theoretical Analysis of XAI Using Post-Hoc Explanation-by-Example from a Survey of ANN-CBR Twin-Systems (2019, methodology): “A review by Gilpin et al. [37] proposes that DNNs have often been explained by using simpler “proxy systems” 4 of which they identify four types: linear models [79], decision trees [84], automatic rule extraction [78], and saliency mapping [85-87].”
Explaining Explanations: An Approach to Evaluating Interpretability of Machine Learning (2018, methodology): “Other approaches aim to reverse engineer the neural network, notably, the RxREN algorithm, which is used in DeepRED[21].”