4. (Total-18%) Consider a binary symmetric channel with input X, output Y, and
transition probability a. More specifically, as shown in the figure below, the input
and output of the channel may be "0" or "1", P(Y=0|X=0)=
P(Y = 1|X = 1) = 1 - a and P(Y = 1|X = 0) = P(Y=0|X=0) = a. The
prior probability is P(X = 0) = p.
(d) (5%) Suppose the prior probability is not known in advance. We manage to
produce an estimated prior probability which may or may not
be equal to the true prior probability P(X = 0) = p. In this case, the cross-
entropy between the true and estimated prior distributions P and is
defined by , which can be
considered as an approximated entropy of X. Please show that the cross-
entropy is always no less than the true entropy H(X) of X, i.e.
(Hint: You may use the Jensen's inequality: plog2 a + (1 -p)log2≤ b <
log2(pa +(1 -p)b) for 0 ≤p ≤ 1, a > 0, and b >0.)
第四題:
某畜產試驗所想調查四種牧草是否影響乳牛產乳量,考量飼養牧場管理差異,選取八個
牧場為區集(block),每個牧場隨機抽取四頭乳牛,並隨機分配餵以四種牧草,並記錄每周產
乳量(單位:公斤),根據所收集資料得到下列的變方分析表(ANOVA):(假設每周產乳量為
常態分布);註:F 分布第 95 個百分位(即 Fα=0.05):F(2,30)=3.32, F(3,30)=2.92,
F(2,20)=3.49, F(3,20)=3.10, F(3,21)=3.07, F(3,31)=2.91, F(8,24)=2.36, F(7,30)=2.33,
F(7,31)=2.32, F(7,20)=2.51, F(7,21)=2.49, F(8,21)=2.42,請回答下列問題: