这篇博客介绍了概率模型的概念，以偏倚硬币为例，解释了伯努利随机变量和独立同分布的原理。接着深入讨论了最大似然估计（MLE），包括似然函数、对数似然函数和求解过程，并提到了似然与概率的关系以及在交叉熵中的应用。最后，简要提及了分类问题中的判别和生成方法。
Probabilistic Models
本lec内容将会在期末考试中占比20%并且是新增内容，在去年的期末试卷中没有出现
A simple biased coin example:
You flip a coin N = 100 N = 100 N=100 times and get outcomes { x 1 , . . . , x N } \{x_1,...,x_N\} { x1,...,xN} where x i ∈ { 0 , 1 } x_i \in \{0,1\} xi∈{ 0,1} and x i = 1 x_i=1 xi=1 is interpreted as heads H H H . Suppose you had N H = 55 N_H =55 NH=55 heads and N T = 45 N_T = 45 NT=45 tails. What is the probability it will come up heads if we flip again? Let’s design a model for this scenario, fit the model. We can use the fit model to predict the next outcome.
The coin is possibly loaded. (这里的loaded指硬币不平衡,也就是正反面的概率不一样) So, we can assume that one coin flip outcome x is a Bernoulli random variable(伯努利随机变量) for some currently unknown parameter θ ∈ [ 0 , 1 ] \theta \in [0,1] θ∈[0,1]
补充: 伯努利随机变量只有两类结果(01,成功失败), 两类结果的可能性之和为1. 也就是
p ( x = 1 ∣ θ ) = θ p(x=1 \mid \theta)=\theta p(x=1∣θ)=θ and p ( x = 0 ∣ θ ) = 1 − θ p(x=0 \mid \theta)=1-\theta p(x=0∣θ)=1−θ
or more succinctly p ( x ∣ θ ) = θ x ( 1 − θ ) 1 − x i p(x \mid \theta)=\theta^x(1-\theta)^{1-x_i} p(
最低0.47元/天 解锁文章
264
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
