Understanding Neural Networks via Rule Extraction

By Rudy Setiono, Huan Liu. 1995

How researchers see this paper

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

Citations in the corpus

Reverse Engineering the Neural Networks for Rule Extraction in Classification Problems (2012, methodology): “Rudy Setiono and Huan Liu have proposed a decompositional technique NeuroLinear [33] which is able to extract oblique classification rules from neural networks with one hidden layer.”
Efficiently Explaining Decisions of Probabilistic RBF Classification Networks (2011, methodology): “Techniques for extracting explicit (if-then) rules from black box models (such as ANN) are described in [11, 7].”
VisualBackProp: efficient visualization of CNNs (2018, methodology): “fying the importance of input variables using a neural network model [27,28], extracting the rules learned by the decision tree model that is ﬁtted to the function learned by the neural network model [29], applying kernel analysis to understand the layer-wise evolution of the representation in a dee”
Rule extraction from neural networks — A comparative study (2012, background): “Rudy Setiono and Huan Liu [13] have proposed a three phase algorithm to understand a neural network via rule extraction.”
VisualBackProp: visualizing CNNs for autonomous driving (2016, methodology): “Finally, understanding CNNs in the context relevant to this paper was also done in the past by characterizing their intriguing and often counter-intuitive properties [25, 26], quantifying the importance of input variables using a neural network model [27, 28], extracting the rules learned by the dec”
Dynamic branch prediction with perceptrons (2001, background): “Techniques have been proposed to extract rules from neural networks [21], but these rules are not always accurate.”
An evolutionary artificial neural networks approach for breast cancer diagnosis (2002, methodology): “In another study, Setiono [26] used his rule extraction from ANNs algorithm [28, 29] to extract useful rules that can predict breast cancer from the Wisconsin dataset.”
Computational intelligence methods for rule-based data understanding (2004, background): “Setiono and Liu [79] use a regularization term in the cost function to itera-”
Pruning algorithms of neural networks — a comparative study (2013, background): “Moreover a small trained network is easier to interpret and the knowledge can be easily extracted in the form of simple rules [10].”
Explainable Artificial Intelligence: a Systematic Review (2020, methodology): “Regarding the decompositional methods, Discretizing Hidden Unit Activation Values by Clustering [237] generates IF-THEN rules by clustering the activation values of hidden neurons and replacing them with the cluster’s average value.”
A Survey on Neural Network Interpretability (2020, methodology): “NeuroRule [51] introduced a three-step procedure of ex-”
Explaining Classifications For Individual Instances (2008, methodology): “In [9] a search-based method has been proposed suitable for ANN with binary units only, while application of the method proposed in [11] requires discretization of the hidden unit activations and pruning of the ANN architecture.”