[go: up one dir, main page]

×
Abstract. Let K(m) denote the smallest number with the property that every m-state finite automaton can be built as a neural net using K(m) or fewer neurons. A ...
The simulation is quite simple: A finite automaton with m states is replaced by a network containing. 2 m + 1 neurons. Part of this research was done while N.
It is proved that a simple algorithm can construct second-order recurrent neural networks with a sparse interconnection topology and sigmoidel discriminant ...
A Low-Energy Implementation of Finite Automata by Optimal-Size Neural Nets ... Ott, Efficient simulation of finite automata by neural nets, Journal of ACM, vol.
Jan 4, 1991 · Abstract. Let K1991 denote the smallest number with the property that every m-state finite automaton can be built as a neural net using K( ...
Abstract. This paper shows the existence of a finite neural network, made up of sigmoidal neurons, which simulates a universal Turing machine.
This paper presents a new architecture of neural networks for representing deterministic finite state automata. The proposed model is capable of strictly ...
People also ask
N. Alon, A.K. Dewdney, T.J. Oft, "Efficient simulation of finite automata by neural nets," J. A.C.M. 38 (1991): 495-514.
This paper presents a new architecture of neural networks for representing deterministic finite state automata (FSA). The proposed model is capable of ...
Many of the constructions rely on common principles such as (1) what it means for a neural network to simulate an automaton; (2) how the mechanisms and the ...