Binary Indexed Trees[二进制索引树]

来源:互联网 发布:装修效果图软件免费 编辑:程序博客网 时间:2024/06/06 09:58

英文原文链接:链接地址

蓝色是笔者注释,高手请忽略

简介

为了使我们的算法更快,我们总是需要一些数据结构。在这篇文章中我们将讨论二进制索引树(Binary Indexed Tree)。依据Peter M. Fenwick,这个数据结构首先用于数据压缩。现在它多用于存储频率和操作累计频率表。


问题定义如下:我们有N个盒子。通常的操作是

1. 在第i个盒子中加入球

2. 求从盒子l到盒子k中球的总和


最天真的做法对于操作1而言时间复杂度是O(1),对于操作2的时间复杂度是O(n)。假设我们查询m次,最坏情况下操作的时间复杂度是O(m*n)。使用一些数据结构(例如RMQ我也不知道是什么东西)可以将这个问题的最差时间复杂度控制在O(m*lg n)。另一种解决方式就是使用Binary Indexed Tree数据结构,最坏情况下的时间复杂度依然是O(m*lg n),然是Binary Indexed Tree更容易编码,也有更小的空间使用量,相比RMQ而言。

注记

BITBinary Indexed Tree 二进制索引树MaxVal maximum value which will have non-zero frequency 非零最大值f[i]frequency of value with index i, i = 1 .. MaxVal 这个可以理解为每个盒子中小球的个数c[i]cumulative frequency for index i (f[1] + f[2] + ... + f[i])tree[i]sum of frequencies stored in BIT with index i (latter will be described what index means); sometimes we will write tree frequency instead sum of frequencies stored in BIT  
在BIT中存储的频率(小球个数)的和;有时在BIT中我们使用tree 频率来替代频率和num¯complement of integer num (integer where each binary digit is inverted: 0 -> 1; 1 -> 0 )  求num的反 NOTE: Often we put f[0] = 0, c[0] = 0, tree[0] = 0, so sometimes I will just ignore index 0.





基本思路

每个整数都可以表示为2的次幂的和。同理,累计的频率也可以表示为子频率集合的和。在我们这篇文章里,每一个集合含有一些连续但互补重合的频率子集。

idx是BIT的索引,r是idx以二进制表示后最右侧的0的位置( 很绕口,解释一下哈。比如idx为12,二进制为1100,则r=2。再来一个,idx=9,二进制1001, 则r=0)。那么tree[idx]是从 ( idx  - 2^ r  + 1)到idx的平率和(看表1.1)(即f[idx - 2^r + 1]+...f[idx])。同时我们还说idx是负责(responsible)从(idx - 2^r + 1)到idx的索引(  注意,这里是算法的关键,也是操作tree的方法)

 12345678910111213141516f1021130425223102c1134588121419212326272729tree1124140122721134029

  table 1.1

(Tips:不要尝试去推理f(i)的值,因为这是给定的例子。c[i]和tree[i]是计算的结果,需要理解)


 12345678910111213141516tree11..231..455..671..899..10119..121313..14151..16

table1.2 responsibility 表 (即tree[i]表示是f[]~f[]的和,例如tree[10]=f[9]+f[10])

image 1.3 tree 负责的index(bar显示的是累加的频率)
 
image 1.4 带有tree频率和的tree

假设我们要寻找index 13的累加频率。在二进制表示中,13表示为1101。据此我们计算c[1101]=tree[1101]+tree[1100]+tree[1000]。

找出最后的1

我们需要多次的从二进制数中获得最后一个1,所以我们需要一个高效的方法。假设我们想从num中获取最后的1.在二进制中num可以表示为a1b,a代表最后一个1之前的所有二进制位,b表示在这个1之后的0.
-num  = (a1b)¯+ 1 = a¯0b¯ + 1.  由于b全市由0构成,所以b¯全部是1.由此可得
-num = (a1b)¯ + 1 = a¯0b¯ + 1 = a¯0(0...0)¯ + 1 = a¯0(1...1) + 1 = a¯1(0...0) = a¯1b.
我们现在可以简单的获得最后一个1,让num和-num做位与运算:

           a1b
&      a¯1b
--------------------
= (0...0)1(0...0)


读取累计的频率和
如果我们需要读取整数idx的频率累计和,我们可以让sum加上tree[idx]的值,然后让idx减去最有一个1(我们也可以说移走最后的1,使最后的1变为0),然后重复上述过程直至idx为0.我们可以使用下面这段代码(C++)。
1int read(int idx){
2    int sum = 0;
3    while (idx > 0){
4        sum += tree[idx];
5        idx -= (idx & -idx);
6    }
7    return sum;
8}

举例: idx=13,sum=0:


iterationidxposition of the last digitidx & -idxsum113 = 110101 (2 ^0)3212 = 110024 (2 ^2)1438 = 100038 (2 ^3)2640 = 0---------

 
image 1.5 箭头指示了在遍历过程中使用的数据.

所以我们的结果是26.这个函数中遍历的次数是idx含有的1的个数,最大的便利次数也就是log MaxVal。
时间复杂度: O(log MaxVal).
代码的复杂度: 如上代码

改变一些位置的频率并更新tree

当改变某些位置的频率时,所有tree中负责该位置的都需要更新。在读取idx的累计和时我们移走idx最后的1并且循环继续。修改tree中的一些值val时,我们需要增加当前idx的tree值tree[idx],增加idx最后一位的1(例如idx为6,该值增加了val,当tree[6]增加了val后。6的最后等于1的一位是2,所以6+2=8,需要继续修改tree[8]的值)并且循环继续之前的过程,只要idx小于MaxVal.C++写的函数如下
1void update(int idx ,int val){
2    while (idx <= MaxVal){
3        tree[idx] += val;
4        idx += (idx & -idx);
5    }
6}
例如idx=5:
iterationidxposition of the last digitidx & -idx15 = 10101 (2 ^0)26 = 11012 (2 ^1)38 = 100038 (2 ^3)416 = 10000416 (2 ^4)532 = 100000------
image 1.6 更新idx为5的频率时遍历的顺序
使用如上算法我们可以更新整个BIT。
时间复杂度: O(log MaxVal)
代码长度:最长10行

读取某个位置的频率值

(未翻译)

整个树乘以或除以某个常数

(未翻译)

给定累计的频率值,找出index(翻译这么多终于到我要用的地方了)

最笨最天真的解决方法就是遍历整个索引,计算累计频率,检查是否等于给定的值。如果考虑存在负数的话,这是唯一解决方案。但是如果我们只有非负的频率值的话(也就是说对于递增的index,累计频率值不减少)我们可以找到指数级的算法,这个算法由二分搜索修改而来。逐步遍历所有的位(从最高为开始),比较当前index的累计频率和给出的值,依据大于小于结果选择高一半或者低一半(就像二分查找)。C++写的函数如下:
01// if in tree exists more than one index with a same
02// cumulative frequency, this procedure will return
03// some of them (we do not know which one)
04 
05// bitMask - initialy, it is the greatest bit of MaxVal
06// bitMask store interval which should be searched
07int find(int cumFre){
08    int idx = 0; // this var is result of function
09     
10    while ((bitMask != 0) && (idx < MaxVal)){ // nobody likes overflow :)
11        int tIdx = idx + bitMask; // we make midpoint of interval
12        if (cumFre == tree[tIdx]) // if it is equal, we just return idx
13            return tIdx;
14        else if (cumFre > tree[tIdx]){
15                // if tree frequency "can fit" into cumFre,
16                // then include it
17            idx = tIdx; // update index
18            cumFre -= tree[tIdx]; // set frequency for next loop
19        }
20        bitMask >>= 1; // half current interval
21    }
22    if (cumFre != 0) // maybe given cumulative frequency doesn't exist
23        return -1;
24    else
25        return idx;
26}
27 
28 
29 
30// if in tree exists more than one index with a same
31// cumulative frequency, this procedure will return
32// the greatest one
33int findG(int cumFre){
34    int idx = 0;
35     
36    while ((bitMask != 0) && (idx < MaxVal)){
37        int tIdx = idx + bitMask;
38        if (cumFre >= tree[tIdx]){
39                // if current cumulative frequency is equal to cumFre,
40                // we are still looking for higher index (if exists)
41            idx = tIdx;
42            cumFre -= tree[tIdx];
43        }
44        bitMask >>= 1;
45    }
46    if (cumFre != 0)
47        return -1;
48    else
49        return idx;
50}

当cumFre为21 时调用find的情况:

First iterationtIdx is 16; tree[16] is greater than 21; half bitMask and continueSecond iterationtIdx is 8; tree[8] is less than 21, so we should include first 8 indexes in result, remember idx because we surely know it is part of result; subtract tree[8] of cumFre (we do not want to look for the same cumulative frequency again - we are looking for another cumulative frequency in the rest/another part of tree); half bitMask and contiueThird iterationtIdx is 12; tree[12] is greater than 9 (there is no way to overlap interval 1-8, in this example, with some further intervals, because only interval 1-16 can overlap); half bitMask and continueForth iterationtIdx is 10; tree[10] is less than 9, so we should update values; half bitMask and continueFifth iterationtIdx is 11; tree[11] is equal to 2; return index (tIdx)
时间复杂度: O(log MaxVal)
代码长度: 小于20行
0 0
原创粉丝点击