You wish, but the reality says no.
In the paper they look at 14, 20, and 32 bit elliptic curves. The corresponding weights storage is 2
13, 2
13.5, 2
14.2. Theoretically the storage would be in the order of 2
7, 2
10, and 2
16. So all looks good, no breakthrough.
To even learn the secp256k1 weights you'd need at least 2
128 examples. Good luck executing that.
I think small neural networks can not handle with Secp256k1 success problem curve with large number it make very complex
neural networks is small digit work with neurons
problem is still on large number
other idea is create some algorithm to predict will be small and easy than may be correct at 50% only can call success