自回归模型索引帖
#refplus, #refplus li{
padding:0;
margin:0;
list-style:none;
}；
document.querySelectorAll(".refplus-num").forEach((ref) => {
let refid = ref.firstChild.href.replace(location.origin+location.pathname,'');
let refel = document.querySelector(refid);
let refnum = refel.dataset.num;
let ref_content = refel.innerText.replace(`[${refnum}]`,'');
tippy(ref, {
content: ref_content,
...
自回归模型概览
【摘要】
【原文】 Murphy, Kevin P. Chapter 22 of Probabilistic Machine Learning: Advanced Topics. MIT Press, 2023. probml.ai.
【参考】
PixelCNN++: Improving the PixelCNN with Discretized Logistic Mi\boldsymbol{x}_{t}ure Likelihood and Other Modifications
https://github.com/openai/pixel-cnn
1 概述
根据概率链式法则，我们可以写出 T 个变量上的任意联合分布如下：
p(x1:T)=p(x1)p(x2∣x1)p(x3∣x2,x1)p(x4∣x3,x2,x1)…=∏t=1Tp(xt∣x1:t−1)p\left(\boldsymbol{x}_{1: T}\right)=p\left(\boldsymbol{x}_1\right) p\left(\boldsymbol{x}_2 \mid \boldsymbo ...