HIRE-ASSISTANT (n) 1 best =0 /candidate 0 is a least-qualified dummy candidate 2 for i 1ton 3 interview candidate i 4 if candidate i is better than candidate best 5 best i 6 hire candidate i 问题4: 在hiring problem算法中假设应 聘者随机到达”意味着什么?
HIRE-ASSISTANT(n) 1 best =0 /candidate 0 is a least-qualified dummy candidate 2 fori 1ton 3 interview candidate i 4 if candidate i is better than candidate best 5 best i 6 hire candidate i 问题5: 在niring problem算法中分析算 法所需要的随机变量是什么?
Hiring-Assistant算法的平均情况分析 涉及的随机变量: 。Hiring操作执行次数:X E[X]=∑xPr{X=x t= 这个公式的计算,难在哪里? 观察到这个问题其实是在分析(甚至是统计)重 复独立试验,能否简化?
Hiring-Assistant算法的平均情况分析 涉及的随机变量: • Hiring操作执行次数:X 这个公式的计算,难在哪里? 观察到这个问题其实是在分析(甚至是统计)重 复独立试验,能否简化?
Indicator Random variable 1 if A occurs, IA0if A does mot occur Indicator random variables provide a convenient method for converting between probabilities and expectations. E[X]=∑Pr{X=x E=P(I=1)=P(the event occurs)
Indicator Random Variable E(I) = P(I = 1) = P(the event occurs)