Given that recurrent neural networks are equivalent to a Turing machine, then why isn't the evolutionary Turing machine, e.g. described in the paper Evolution of evolution: Self-constructing Evolutionary Turing Machine case study (2007), mainstream?
$\begingroup$
$\endgroup$
-
2$\begingroup$ The main problem with genetic algorithm is, that in the naive understanding they are generating a state space which is to large to compute on mainstream hardware. If the algorithm is extended with heuristics to a neural symbolic machine, it might become a useful program solving technique, but these additions are complex to describe. $\endgroup$ – Manuel Rodriguez Jul 22 '19 at 7:10
