i-RevNet: Deep Invertible Networks

Jörn-Henrik JacobsenArnold SmeuldersEdouard Oyallon

   Papers with code   Abstract  PDF

It is widely believed that the success of deep convolutional networks is based on progressively discarding uninformative variability about the input with respect to the problem at hand. This is supported empirically by the difficulty of recovering images from their hidden representations, in most commonly used network architectures... (read more)

Benchmarked Models

RANK
MODEL
REPO
CODE RESULT
PAPER RESULT
ε-REPRODUCED
BUILD
1
i-RevNet-301
74.0%
--