We introduce a novel approach for parameter generation, named neural network parameter diffusion (p-diff), which employs a standard latent diffusion model to synthesize a new set of parameters
828
stars
42
forks
source link
Whether using BN in the part "Generalization on entire model parameters". #12
In the experiment of "Generalization on entire model parameters', you mentioned two small architectures, CNN-3 and MLP-3, and I am not sure whether there are any BN following every layer.
Hi,
In the experiment of "Generalization on entire model parameters', you mentioned two small architectures, CNN-3 and MLP-3, and I am not sure whether there are any BN following every layer.
Thank you.