Classifier-Free Gaussian Guidance
Check when an extrapolated score is the gradient of a normalizable density.
Free theory articleCheck when an extrapolated score is the gradient of a normalizable density.
Concept & interpretation
Classifier-free guidance extrapolates between an unconditional score and a conditional score to improve sample quality. We analyze when this extrapolated score remains the gradient of a valid, normalizable probability density.
Experiment
Open the Classifier-Free Gaussian Guidance module to observe how guidance weight and variance affect the resulting distribution.
- Set Conditional sigma to "2" and Guidance weight to "1". The Density status is Normalizable.
- Increase Guidance weight to "4/3". The Guided precision drops to 0.0, and the status changes to Not normalizable.
- Change Conditional sigma to "0.5" and Guidance weight to "2". The precision becomes 7.0 (a very sharp density with variance ~0.14).
- Change Probe x to verify the scores at different points while the global density parameters remain the same.