The Induction of Dynamical Recognizers
Machine Learning - Connectionist approaches to language learning
Random DFA's can be approximately learned from sparse uniform examples
COLT '92 Proceedings of the fifth annual workshop on Computational learning theory
Induction of finite-state languages using second-order recurrent networks
Neural Computation
Learning finite machines with self-clustering recurrent networks
Neural Computation
Extraction of rules from discrete-time recurrent neural networks
Neural Networks
Constructing deterministic finite-state automata in recurrent neural networks
Journal of the ACM (JACM)
Self-organizing maps
On the emergence of rules in neural networks
Neural Computation
Fuzzy finite-state automata can be deterministically encoded into recurrent neural networks
IEEE Transactions on Fuzzy Systems
Rule Extraction from Recurrent Neural Networks: A Taxonomy and Review
Neural Computation
The Crystallizing Substochastic Sequential Machine Extractor: CrySSMEx
Neural Computation
Learning symbolic representations of hybrid dynamical systems
The Journal of Machine Learning Research
Hi-index | 0.00 |
Neural networks do not readily provide an explanation of the knowledge stored in their weights as part of their information processing. Until recently, neural networks were considered to be black boxes, with the knowledge stored in their weights not readily accessible. Since then, research has resulted in a number of algorithms for extracting knowledge in symbolic form from trained neural networks. This article addresses the extraction of knowledge in symbolic form from recurrent neural networks trained to behave like deterministic finite-state automata (DFAs). To date, methods used to extract knowledge from such networks have relied on the hypothesis that networks' states tend to cluster and that clusters of network states correspond to DFA states. The computational complexity of such a cluster analysis has led to heuristics that either limit the number of clusters that may form during training or limit the exploration of the space of hidden recurrent state neurons. These limitations, while necessary, may lead to decreased fidelity, in which the extracted knowledge may not model the true behavior of a trained network, perhaps not even for the training set. The method proposed here uses a polynomial time, symbolic learning algorithm to infer DFAs solely from the observation of a trained network's input-output behavior. Thus, this method has the potential to increase the fidelity of the extracted knowledge.