A Machine Learning Method for Extracting Symbolic Knowledge from Recurrent Neural Networks

By Anwar Vahed, Christian Walter Peter Omlin. 2004

17 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Electrical and Electronic Engineering

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

1. Henrik Jacobsson
2. J. Urgen Schmidhuber
3. David R Ha
4. Alex Graves
5. Sergey Levine
6. Yoshua Bengio
7. Yoav Goldberg
8. Quoc V. Le

In Electrical and Electronic Engineering

1. J. Urgen Schmidhuber
2. Henrik Jacobsson
3. Rupesh Kumar Srivastava

Citations in the corpus

• Deep learning in neural networks: An overview (2014, methodology): “LSTM clearly outperformed previous RNNs on tasks that require learning the rules of regular languages describable by deterministic Finite State Automata (FSAs) (Watrous and Kuhn, 1992; Casey, 1996; Siegelmann, 1992; Blair and Pollack, 1997; Kalinke and Lehmann, 1998; Zeng et al., 1994; Manolios and ”
• Rule Extraction from Recurrent Neural Networks: ATaxonomy and Review (2005, methodology): “One simple approach is to avoid the problem by not modelling transitions at all (Tiˇno et al. 2004), or even not to quantise the state space at all (Vahed & Omlin 1999, Vahed & Omlin 2004).”