Re:Tversky Neural Networks
by sdenton4 on Sunday August 17, 2025 @08:30AM
(#44931149)
It's wishful thinking.
Neutral networks need to be over parameterized to find good solutions, meaning there is a surface of solutions. The optimization procedure tries to walk towards that surface as quickly as possible, and tend to find a low-energy point on the surface of solutions. In particular, a low energy solution isn't sparse, and therefore isn't interpretable.
Neutral networks need to be over parameterized to find good solutions, meaning there is a surface of solutions. The optimization procedure tries to walk towards that surface as quickly as possible, and tend to find a low-energy point on the surface of solutions. In particular, a low energy solution isn't sparse, and therefore isn't interpretable.