Formal languages
Constructing deterministic finite-state automata in recurrent neural networks
Journal of the ACM (JACM)
Introduction To Automata Theory, Languages, And Computation
Introduction To Automata Theory, Languages, And Computation
Supervised neural networks for the classification of structures
IEEE Transactions on Neural Networks
A general framework for adaptive processing of data structures
IEEE Transactions on Neural Networks
On the computational power of Elman-style recurrent networks
IEEE Transactions on Neural Networks
Encoding Nondeterministic Finite-State Tree Automata in Sigmoid Recursive Neural Networks
Proceedings of the Joint IAPR International Workshops on Advances in Pattern Recognition
Neural Networks for Adaptive Processing of Structured Data
ICANN '01 Proceedings of the International Conference on Artificial Neural Networks
Recursive self-organizing network models
Neural Networks - 2004 Special issue: New developments in self-organizing systems
Rule Extraction from Recurrent Neural Networks: A Taxonomy and Review
Neural Computation
Recurrent networks for structured data - A unifying approach and its properties
Cognitive Systems Research
Hi-index | 0.00 |
Recently, a number of authors have explored the use of recursive neural nets (RNN) for the adaptive processing of trees or tree-like structures. One of the most important language-theoretical formalizations of the processing of tree-structured data is that of deterministic finite-state tree automata (DFSTA). DFSTA may easily be realized as RNN using discrete-state units, such as the threshold linear unit. A recent result by Síima (Neural Network World7 (1997), pp. 679驴686) shows that any threshold linear unit operating on binary inputs can be implemented in an analog unit using a continuous activation function and bounded real inputs. The constructive proof finds a scaling factor for the weights and reestimates the bias accordingly. In this paper, we explore the application of this result to simulate DFSTA in sigmoid RNN (that is, analog RNN using monotonically growing activation functions) and also present an alternative scheme for one-hot encoding of the input that yields smaller weight values and, therefore, works at a lower saturation level.