Oral
Invertible Residual Networks
Jens Behrmann · Will Grathwohl · Tian Qi Chen · David Duvenaud · Joern-Henrik Jacobsen

Wed Jun 12th 02:00 -- 02:20 PM @ Hall A

We show that standard ResNet architectures can be made invertible, allowing the same model to be used for classification, density estimation, and generation. Typically, enforcing invertibility requires partitioning dimensions or restricting network architectures. In contrast, our approach only requires adding a simple normalization step during training, already available in standard frameworks. Invertible ResNets define a generative model which can be trained by maximum likelihood on unlabeled data. To compute likelihoods, we introduce a tractable approximation to the Jacobian log-determinant of a residual block. Our empirical evaluation shows that invertible ResNets perform competitively with both state-of-the-art image classifiers and flow-based generative models, something that has not been previously achieved with a single architecture.

Author Information

Jens Behrmann (University of Bremen)
Will Grathwohl (University of Toronto)
Ricky T. Q. Chen (U of Toronto)
David Duvenaud (University of Toronto)
Jörn Jacobsen (Vector Institute and University of Toronto)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors