In the Deep Equilibrium Model the neural network can be seen as "infinitely deep". Training learns a nonlinear function as usual. But there is no forward propagation of input data through layers. Instead, a root finding problem is solved when data comes in.

My question is, what is actually the function for which roots are searched, I'm struggling to see what would be unknown when data is available and parameters have been found in training?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.