推荐GaggiX · 2023-03
“also: Cold Diffusion, to see the diffusion process applied to processes other than Gaussian noise.”
推荐它,可以看到扩散过程用在高斯噪声以外的变换上
证明用模糊、遮挡等确定性退化也能构造扩散式生成模型
作者观察到扩散模型的生成能力并不强依赖于加高斯噪声这一种退化方式,改用模糊、掩码等完全确定性的退化,同样能沿用扩散模型的训练与测试更新规则得到生成模型。这一结果动摇了扩散模型必须依赖噪声(朗之万动力学或变分推断)的主流解释,并指向可逆转任意过程的广义扩散模型。
HN 上的人引用它说明扩散过程不必是高斯噪声,原论文的许多假设可以去掉,模型依然有效。
“also: Cold Diffusion, to see the diffusion process applied to processes other than Gaussian noise.”
“I think it might be this paper [1] succintly described by the author in this twitter thread [2]”
“The explanation in the original paper turns out not to be true; you can get rid of most of their assumptions and it still works: Cold Diffusion”