Evolution-based training for neurgo, a neural network framework in Go.
- Learning via Stochastic Hill Climbing works
- Topological mutatation operators implemented
- Example which evolves a network capable of solving XOR
$ cd examples
$ go build -v && go run run_examples.go run_stochastic_hill_climber.go
DXNN2 - Pure Erlang TPEULN (Topology & Parameter Evolving Universal Learning Network).
Handbook of Neuroevolution Through Erlang by Gene Sher.