On the combat side the top games are War Thunder (which I don't recommend due to it's microtransaction model), DCS World (flawed and expensive but still fun), IL-2 Battle of Stalingrad (good WW2 game) and Falcon BMS (excellent, but focused on the F-16 and has a small community)
That's useless in this case. You need to be able to prove that it will work with all inputs, and there are too many combinations of inputs to exhaustively enumerate.
There is no way to determine that a non-trivial neural network won't drastically diverge in output due to small changes in input (eg one pixel attacks on image classifiers). This is true for all current models I know of.
Almost all neural network implementations have continuous outputs (ie the nodes in the output layer produce a value between 0 and 1). That doesn't change the above issue at all.
This is much less of an issue with traditional methods
Not quite. Lifetime annotations prevent you from accidentally using references after the value they refer to has been freed. They track how long things will live, instead of defining how long they will live.
Basically they turn use-after-free errors into compile errors.
(I'm using 'free' here to mean cleaned up in general. Lifetimes can track stack values.)
Not the issue here, look at the end of the selected path, where the cursor is. If you go down where the cursor is instead of right the path is 1 step shorter.
There is no line. If Google decides to ban you, your only recourse is to post online and hope your story gets enough traction for it to become a PR issue for them.
I don't care if they 'dragnet survey' my users who are just looking for 20-year-old data sheets. Fact is, not everything needs unbreakable encryption, or a warning label for lacking it.
A lot of people also seem to like VTOL VR: https://store.steampowered.com/app/667970/VTOL_VR/