← Search

Jakob N Foerster

2 accepted papers

2017

Input Switched Affine Networks: An RNN Architecture Designed for Interpretability

ICML 2017poster

There exist many problem domains where the interpretability of neural network models is essential for deployment. Here we introduce a recurrent architecture composed of input-switched affine transformations – in other words an RNN without any explicit nonlinearities, but with input-dependent recurre…

Cited by 41SourcePDFScholar