Term
gradient descent
A way to train a neural network: after each answer, all the weights are nudged slightly in the direction that makes the error smaller, and this is repeated an enormous number of times.
If Anyone Builds It, Everyone Dies: Why Superhuman AI Would Kill Us All, Eliezer Yudkowsky, Nate Soares
Explained in
- If Anyone Builds It, Everyone Dies: Why Superhuman AI Would Kill Us All, Eliezer Yudkowsky, Nate Soares
- What Is ChatGPT Doing, Stephen Wolfram
A way of training: the weights are nudged little by little in the direction where the error gets smaller, like a hiker walking down a slope in the fog.
