Machine Learning Innovation – It would be like if a machine learning iteration is similar in method to calculating a differential equation. Sometime you take very small steps. Like A Stiff Machine Learning Problem? Interpolation of iterations?

Sorry this was not how I remember the tweet.

Anyway. I wonder. If you have a stiff problem. One that requires a lot of small time steps. Could this be made into Random Loss Problem. Where use GAN Questions 0/1 somehow and get rid of a lot of loss function checks and updates.