Reverse Engineering the Neural Networks for Rule Extraction in Classification Problems

By M. Augasta, Thangairulappan Kathirvalavakumar. 2011

Subfields: 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 Safety, Robustness & Interpretability

Citations in the corpus

Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AI (2019, neutral): “• Some methods [82,144] are classiﬁed on a single category ( Explanation by simpliﬁcation for Multi-Layer Neural Network ) in Fig.”
Explaining Explanations: An Overview of Interpretability of Machine Learning (2018, methodology): “DeepRED utilizes several strategies to simplify its decision trees: it uses RxREN [23] to prune unnecessary input, and it applies algorithm C4.5 [24], a statistical method for creating a parsimonious decision tree.”
Explaining Explanations: An Approach to Evaluating Interpretability of Machine Learning (2018, methodology): “DeepRED utilizes several strategies to simplify its decision trees: it uses RxREN [23] to prune unnecessary input, and it applies algorithm C4.5 [24], a statistical method for creating a parsimonious decision tree.”
Deep Logic Networks: Inserting and Extracting Knowledge From Deep Belief Networks (2018, unknown): “Knowledge extraction after training is expected to provide explanations [10]–[14], as already mentioned.”
Principles and Practice of Explainable Machine Learning (2020, methodology): “Augusta and Kathirvalavakumar [4] have introduced the RxREN algorithm, employing reverse engineering techniques in order to analyse the output and trace back the components that cause the ﬁnal result.”
Explainable Artificial Intelligence: a Systematic Review (2020, background): “Rule Extraction by Reverse Engineering (RxREN) [242] relies on a reverse engineering technique to trace back input neurons that cause the final result, whilst pruning the insignificant ones, and to determine the data ranges of each significant neuron in respective classes.”
Trustworthy AI: From Principles to Practices (2021, methodology): “Various approaches have been proposed to approximate ML models, such as random forests [317, 392] and neural networks [28, 86, 393].”
Pruning algorithms of neural networks — a comparative study (2013, background): “Moreover a simple and comprehensible set of rules can be extracted from the network by removing the input neurons that are not mandatory for solving the problem [13] [14].”