Euclidean Norm Regularization

Introduced by Wu et al. in Deep Compressed Sensing

Euclidean Norm Regularization is a regularization step used in generative adversarial networks, and is typically added to both the generator and discriminator losses:

$$ R_{z} = w_{r} \cdot ||\Delta{z}||^{2}_{2} $$

where the scalar weight $w_{r}$ is a parameter.

Image: LOGAN

Source: Deep Compressed Sensing

Latest Papers

PAPER DATE
LOGAN: Latent Optimisation for Generative Adversarial Networks
| Yan WuJeff DonahueDavid BalduzziKaren SimonyanTimothy Lillicrap
2019-12-02
Deep Compressed Sensing
| Yan WuMihaela RoscaTimothy Lillicrap
2019-05-16

Tasks

TASK PAPERS SHARE
Conditional Image Generation 1 33.33%
Image Generation 1 33.33%
Meta-Learning 1 33.33%

Components

COMPONENT TYPE
🤖 No Components Found You can add them if they exist; e.g. Mask R-CNN uses RoIAlign

Categories