> Considering speed:
> Reusing of old information like Bradley and I have proposed might
> drastically increase the speed.
I do not see which discussion you are referring too.
The gradient algorithm only updates areas that need updating as it starts from
goal and expands.
Maybe we could add some heuristics to prune propagation but I'm far from
convinced that it will help.
I a while back wrote a thread which discussed ways a gradient could be updated, rather than fully recomputed. This could save magnitudes of time alone. I also suggested several other gradient speed ups.