Latent Optimisation is a technique used for generative adversarial networks to refine the sample quality of $z$. Specifically, it exploits knowledge from the discriminator $D$ to refine the latent source $z$. Intuitively, the gradient $\nabla_{z}f\left(z\right) = \delta{f}\left(z\right)\delta{z}$ points in the direction that better satisfies the discriminator $D$, which implies better samples. Therefore, instead of using the randomly sampled $z \sim p\left(z\right)$, we uses the optimised latent:
$$ \Delta{z} = \alpha\frac{\delta{f}\left(z\right)}{\delta{z}} $$
$$ z' = z + \Delta{z} $$
Source: LOGAN .
Source: Deep Compressed SensingPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Image Generation | 3 | 12.00% |
Bias Detection | 2 | 8.00% |
Clustering | 2 | 8.00% |
Medical Image Generation | 2 | 8.00% |
Conditional Image Generation | 2 | 8.00% |
Fairness | 1 | 4.00% |
Edge-computing | 1 | 4.00% |
Texture Synthesis | 1 | 4.00% |
Anatomy | 1 | 4.00% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |