Lifted Regression/Reconstruction Networks
Paper in proceeding, 2020

In this work we propose lifted regression/reconstruction networks(LRRNs), which combine lifted neural networks with a guaranteed Lipschitz continuity property for the output layer. Lifted neural networks explicitly optimize an energy model to infer the unit activations and therefore—in contrast to standard feed-forward neural networks—allow bidirectional feedback between layers. So far lifted neural networks have been modelled around standard feed-forward architectures. We propose to take further advantage of the feedback property by letting the layers simultaneously perform regression and reconstruction. The resulting lifted network architecture allows to control the desired amount of Lipschitz continuity, which is an important feature to obtain adversarially robust regression and classification methods. We analyse and numerically demonstrate applications for unsupervised and supervised learning

lifted neural networks

energy-based models

adversarial robustness

Lipschitz continuity

Author

Rasmus Kjær Høier

Chalmers, Electrical Engineering, Signal Processing and Biomedical Engineering

Christopher Zach

Chalmers, Electrical Engineering, Signal Processing and Biomedical Engineering

31st British Machine Vision Conference, BMVC 2020

31st British Machine Vison Conference 2020, BMVC 2020
Virtual conference, United Kingdom,

Subject Categories

Computer Engineering

More information

Latest update

1/25/2024