classifier-free guidance does: interpolates between conditional and unconditional generation

Boosting (machine learning)

Boosting reduces bias in ML models

score matching does: learns the gradient of the log-density without normalizing

Matching score learns gradient of log-density without normalizing

Adam has bias correction: divides by (1-β^t) in early steps

Adam bias correction divides by (1-β^t) in early steps to counteract initial bias from accumulated gradients

Adam vs SGD: Adam adapts per-parameter rates, SGD often generalizes better with tuning

Adam adjusts learning rates per-parameter, SGD generalizes better with tuning

to use random forests: when you want a strong baseline with minimal hyperparameter tuning

Random forests are ideal for robust baseline models with minimal hyperparameter tuning

denoising score matching does: learns to denoise, which equals learning the score

Denoising score matching learns to denoise by estimating the score (gradient of log probability) of data distributions