记录一下概念(信息论中)

记录一些常见的概念。

(香侬)熵

Information entropy is the average rate at which information is produced by a stochastic source of data.

\[H[x]=-\sum_{x} p(x) \log _{2} p(x) \]

更多:https://en.wikipedia.org/wiki/Entropy_(information_theory)

交叉熵

\[{\displaystyle H(p,q)=-\sum _{x\in {\mathcal {X}}}p(x)\,\log q(x)= {\displaystyle H(p)+D_{\mathrm {KL} }(p\|q)}} \]

更多:https://en.wikipedia.org/wiki/Cross_entropy

Kullback–Leibler divergence

也叫相对熵(relative entropy)

\[{\displaystyle D_{\text{KL}}(P\parallel Q)=\int _{-\infty }^{\infty }p(x)\log \left({\frac {p(x)}{q(x)}}\right)\,dx} \]

  • 非对称
  • 非负
  • 凸性

更多: https://en.wikipedia.org/wiki/Kullback–Leibler_divergence

Jensen–Shannon divergence

也叫information radius,total divergence to the average

\[{\displaystyle {\rm {JSD}}(P\parallel Q)={\frac {1}{2}}D(P\parallel M)+{\frac {1}{2}}D(Q\parallel M)} \]

其中\({\displaystyle M={\frac {1}{2}}(P+Q)}\)

  • 有界
  • 对称

更多:https://en.wikipedia.org/wiki/Jensen–Shannon_divergence

f-divergence

\[{\displaystyle D_{f}(P\parallel Q)\equiv \int _{\Omega }f\left({\frac {dP}{dQ}}\right)\,dQ.} \]

更多:https://en.wikipedia.org/wiki/F-divergence

posted @ 2019-03-30 17:13  huiwong  阅读(262)  评论(0编辑  收藏  举报