Re:Can a Linear Program Do Weight-Space Programming?
by quietbritishjim on Thursday November 12, 2026 @04:03PM
(#50474242)
No.
The neutral network was originally produced by training it in the usual way. So the training data is the "source code" and back propagation / gradient descent optimiser is the "compiler". That's what the human used to produce the network.
The weights are analogous to the compiled machine code. The LLM bypassed the compiler and edited those directly.
The neutral network was originally produced by training it in the usual way. So the training data is the "source code" and back propagation / gradient descent optimiser is the "compiler". That's what the human used to produce the network.
The weights are analogous to the compiled machine code. The LLM bypassed the compiler and edited those directly.