2017 (modified: 11 Nov 2022)ICML 2017Readers: Everyone
Abstract:This paper presents the input convex neural network architecture. These are scalar-valued (potentially deep) neural networks with constraints on the network parameters such that the output of the n...