I hope they checked their work, wouldn't want someone to pick holes in it.
Neural networks whip fleshbag butt at identifying craters
A neural network can wipe the floor with fleshy researchers at that most tedious of cosmic tasks – spotting craters. Eyeballing craters is usually done manually, which tends to result in only the largest impacts being spotted, or via a crater detection algorithm (CDA), which works well on data it has been trained on but gets a …
COMMENTS
-
This post has been deleted by its author
-
-
Friday 16th March 2018 18:26 GMT Anonymous Coward
A bit confused?
"...via a crater detection algorithm (CDA), which works well on data it has been trained on but gets a bit stroppy when shown something new."
Getting "a bit stroppy when shown something new" is a limitation of neural networks - claiming that using a neural network will overcome this problem doesn't make sense.
On the other hand, actually using a CNN for crater identification makes lots of sense - it's the sort of thing CNNs can be really good at, and certainly better than a simple procedure based algorithm.
It's just the justification that seems a little confused.
-