swqeaaa的博客

ActivationTime Limit: 20000/10000 MS (Java/Others)Memory Limit: 32768/32768 K (Java/Others)Total Submission(s): 4808Accepted Submission(s): 1599Problem DescriptionAfter 4 years' waiti...

Aeroplane chessTime Limit: 2000/1000 MS (Java/Others)Memory Limit: 32768/32768 K (Java/Others)Total Submission(s): 5705Accepted Submission(s): 3562Problem DescriptionHzz loves aeropla...

Collecting BugsTime Limit:10000MS Memory Limit:64000K Total Submissions:7554 Accepted:3586 Case Time Limit:2000MS Special Judge DescriptionIvan is fond of collec...

题意: 一个武器共有n+1级, 给出第i级升到i+1级的花费, 成功概率, 失败掉回第几级, 给出q次询问, 每次问从第l级升到第r级的期望花费.可以发现这个题的期望具有可加性: 设dp[i]dp[i]dp[i]表示iii升到i+1i+1i+1的期望花费, 那么iii升到i+2i+2i+2的花费就是dp[i]+dp[i+1]dp[i]+dp[i+1]dp[i]+dp[i+1], 那么对于任意ll...

给出1到100共100个格子排成一行, 到达一些格子时会瞬间转移, 每次投骰子来确定走的步数, 如果走出格子之外就重新投, 问走到100号格子的期望步数.如果每个格子瞬间转移只会向后的话, 就是普通的期望dp—只需从后往前dp即可. 但是有些格子可能会往回跳, 所以不满足无后效性.按照普通期望dp的做法, dp[i]表示从i这里开始走, 走到终点的期望步数. 可惜在当前情景下, 我们计算i时可...