Are Arti cial Neural Networks Black Boxes ?

By J. M. Ben. 1996

How researchers see this paper

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

Explaining Explanations: An Overview of Interpretability of Machine Learning (2018, background): “There has also been work on transforming neural network to fuzzy rules [30], by transforming each neuron into an approximate rule.”
Ten years of genetic fuzzy systems: current framework and new trends (2001, background): “Fuzzy neural networks contribute to overcome lacking of linguistic representation and transparency, but the designer must decide on major design parameters such as universes granulation, rule antecedent aggregation operators, rule semantics, rule base aggregation operators and defuzzi5cation methods”
Extracting symbolic rules from trained neural network ensembles (2003, unknown): “Ben-itez et al. [3] establish the equality between a certain class of neural networks and fuzzy rule based systems so that the network could be explained by a set of fuzzy rules.”
Explaining Explanations: An Approach to Evaluating Interpretability of Machine Learning (2018, background): “There has also been work on transforming neural network to fuzzy rules [30], by transforming each neuron into an approximate rule.”
Calibration of Pre-trained Transformers (2020, background): “Neural networks have seen wide adoption but are frequently criticized for being black boxes, offering little insight as to why predictions are made (Benitez et al., 1997; Dayhoff and DeLeo, 2001; Castelvecchi, 2016) and making it difﬁcult to diagnose errors at test-time.”