In this video on Deep Gaussian Processes, Neil Lawrence (at 37:30) mentions that thinking of the layers of a neural network in terms of basis functions from Gaussian processes can be viewed as a motivation/explanation for Ioffe/Szegedy's batch normalization. Can someone explain this comment to me? I'm not understanding what he means.


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.