suraj021's blog

By suraj021, 4 years ago, , I was having trouble understanding Bitmask, then I found an unknown pdf about Bitmask on google. I would like to help the beginners like me in understanding Bitmasks and their uses. Here we go :

#### MOTIVATION

Suppose you have a set of objects and you want some way to represent which objects to pick and which ones not to
pick. How do you represent that in in a program? More generally, how do you represent a subest of a set?One way is to use a Map to associate with each object a boolean value indicating whether the object is picked. Alternatively,if the object can be indexed by integers, you can use a boolean array. However, this takes up a lot of memory and can be slow due to the overhead of Map and array. If the size of the set is not too large, a bitmask is much more efficient (and convenient)!

#### WHAT IS BITMASKS ?

Bitmasks a.k.a. lightweight, small sets of Booleans (native support in C/C++/Java). An integer is stored in a computer’s memory as a sequence/string of bits. Thus, we can use integers to represent a lightweight small set of Boolean values. All set operations then involve only the bitwise manipulation of the corresponding integer, which makes it a much more efficient choice when compared with the C++ STL vector, bitset, or set options. Such speed is important in competitive programming.

We know an integer is just a bunch of bits stringed together. The 1st bit will represent whether the 1st object is picked, the 2nd bit will represent whether the 2nd object is picked or not, etc. For example, suppose in a set of 5 objects, we have picked the 1st , 3rd , and 4th object. The bitmask to represent this in binary is 01101 or 13 in decimal (in the notes, the 1 st bit will always be the least significant bit and will always appear at the very right).

1. REPRESENTATION: A 32 (or 64)-bit signed integer for up to 32 (or 64) items. ( To avoid issues with the
two’s complement representation, use a 32-bit/64-bit signed integer to represent bitmasks of up to
30/62 items only, respectively ).

For example:                          5| 4  | 3 | 2 | 1 | 0   <- 0-based indexing from right
32| 16 | 8 | 4 | 2 | 1   <- power of 2
A= 34 (base 10) =   1 | 0  | 0 | 0 | 1 | 0   <- in binary
F | E  | D | C | B | A   <- alternative alphabet label
In the example above,the integer A = 34 or 100010 in binary also represents a small set {1, 5} with a
0-based indexing scheme in increasing digit significance ( or {B, F} using the alternative alphabet
label )because the second and the sixth bits (counting from the right) of A are on ( 1 ).

2. To multiply/divide an integer by 2:
We only need to shift the bits in the integer left/right, respectively.
Notice that the truncation in the shift right operation automatically rounds the division-by-2 down,
e.g. 17/2  = 8.

For example:         A = 34 (base 10)                  = 100010 (base 2)
A = A << 1 = A * 2 = 68 (base 10) = 1000100 (base 2)
A = A >> 2 = A / 4 = 17 (base 10) = 10001 (base 2)
A = A >> 1 = A / 2 = 8 (base 10) = 1000 (base 2) <- LSB( Least Significant Bit )is gone

3. Add the jth object to the subset (set the jth bit from 0 to 1):
use the bitwise OR operation A |= (1 << j).

For example:     A = 34 (base 10) = 100010 (base 2)
j = 3, 1 << j    = 001000 <- bit ‘1’ is shifted to the left 3 times
-------- OR (true if either of the bits is true)
A = 42 (base 10) = 101010 (base 2) // update A to this new value 42

4. Remove the jth object from the subset (set the jth bit from 1 to 0):
use the bitwise AND operation A &= ∼(1 << j).

For example:         A = 42 (base 10) = 101010 (base 2)
j = 1, ~(1 << j) = 111101 <- ‘~’ is the bitwise NOT operation
-------- AND
A = 40 (base 10) = 101000 (base 2) // update A to this new value 40

5. Check whether the jth object is in the subset (check whether jth bit is 1):
use the bitwise AND operation T = A & (1 << j).
If T = 0, then the j-th item of the set is off.
If T != 0 (to be precise, T = (1 << j)), then the j-th item of the set is on.

For example:    A = 42 (base 10) = 101010 (base 2)
j = 3, 1 << j    = 001000 <- bit ‘1’ is shifted to the left 3 times
-------- AND (only true if both bits are true)
T = 8 (base 10)  = 001000 (base 2) -> not zero, the 3rd item is on

6. To toggle (flip the status of) the j-th item of the set:
use the bitwise XOR operation A ∧ = (1 << j).

For example:       A = 40 (base 10) = 101000 (base 2)
j = 2, (1 << j)  = 000100 <- bit ‘1’ is shifted to the left 2 times
-------- XOR <- true if both bits are different
A = 44 (base 10) = 101100 (base 2) // update A to this new value 44

7. To get the value of the least significant bit that is on (first from the right):
use T = (A & (-A)).

For example:     A =  40 (base 10) = 000...000101000 (32 bits, base 2)
-A = -40 (base 10) = 111...111011000 (two’s complement)
----------------- AND
T =   8 (base 10) = 000...000001000 (3rd bit from right is on)

8. To turn on all bits in a set of size n: (be careful with overflows)
use A = (1 << n) - 1 ;

9. Iterate through all subsets of a set of size n:
for ( x = 0; x < (1 << n); ++x )

10. Iterate through all subsets of a subset y (not including empty set):
for ( x = y; x > 0; x = ( y & (x-1) ) )

Example of a subset problem: given a set of numbers, we want to find the sum of all subsets.

Sol: This is easy to code using bitmasks. we can use an array to store all the results.

int sum_of_all_subset ( vector< int > s ){
int n = s.size() ;
int results[ ( 1 << n ) ] ;     // ( 1 << n )= 2^n

//initialize results to 0
memset( results, 0, sizeof( results ) ) ;

// iterate through all subsets

for( int i = 0 ; i < ( 1 << n ) ; ++ i ) {    // for each subset, O(2^n)
for ( int j = 0; j < n ; ++ j ) {       // check membership, O(n)
i f ( ( i & ( 1 << j ) ) ! = 0 )    // test if bit ‘j’ is turned on in subset ‘i’?
results[i] += s [j] ;          // if yes, process ‘j’
}
}
}

11. LIMITATIONS:
a. Always check the size of the set to determine whether to use an int or long long or not using bitmask at all
b. Always use parenthesis to indicate the precedence of operations when doing bitwise operations!
When it involves bitwise operators and not putting parenthesis can yield undesirable results!

For example, let x = 5. Then x - 1 << 2 = 16, but x - (1 << 2) = 1

P.S 1. I apologize for bad formatting. 2. If you find something wrong/inappropriate please correct me. 3. Examples are copied from some text book 4. Can anyone please write a blog on Backtracking , i don't get flow of control in recursive calls in backtracking when a certain constraint fails on a configuration ( like in N queens problem ), how does program backtracks and how control flow takes place then and what happens after that ?

Thank You ;  Comments (34)
 » Can there be anything better than exponential in subset sum ?
•  » » 3 years ago, # ^ | ← Rev. 2 →   There is a simple formula for the sum of all subsets. Consider a set {a1, a2, ..., an}. There are 2n subsets, and each ai exists in exactly 2n - 1 of them. Therefore, the sum of all subsets is equal to 2n - 1(a1 + a2 + ... + an).
•  » » » Can you please explain the fact how each ai exists in 2^(n-1) subsets. Thanks
•  » » » » For each ai you have n-1 elements. Each of them can exist and not exist in some subset, so there are 2^(n — 1) such subsets.
•  » » Better use Dynamic Programming. One of the classic question of Dynamic Programming. Subset Sum problem.
 » SEE this link---> beneficial to all
 » Copy & paste from Competitive programming book :D
•  » » I have already written that in P.S :D
•  » » And which book might that be?
•  » » » Just google "Competitive Programming 3". Best of Luck
•  » » » » You seems to know which book it comes from, why not replace in your post "some text book" by the title, and making it more visible? Stealing ideas like this is a ugly thing to do
 » thanks ,such type of activity really helps beginners like me.
 » I think this type of post is illegal
•  » » Look, you have an option of ignoring the post. Just do it. By the way JAYPEE found it useful. I helped a person. :P
 » Hello, How to make sure that at all times, there are only k bits set to 1 in an n bit array?
•  » » @chari407 What operation do you want to do with that n bit array,it depends upon that,after what manipulations you want to have the number of set bits same?
 » Good WORK DUDE
 » Can someone explain this line for me ? results[i] += s [j] ; // if yes, process ‘j’i can't figure what it is doing i exactly and cant figure the relation between if jth bit is turned on in "i" with the array itself.
•  » » There is in i for every possible subset of the set. As in, for every possible configuration of numbers to be added together, there will be one position in the results array. That line adds the j-th number to the i-th subset, if that subset were to contain said j-th number, example:/Initial set = {1, 2, 3}, possible subsets are 8:111 = 1 + 2 + 3 = 6 110 = 1 + 2 = 3 101 = 1 + 3 = 4 100 = 1 = 1 011 = 2 + 3 = 5 010 = 2 = 2 001 = 3 = 3 000 = 0 = 0The final results will be the results stored in the 8 positions of the result array, the bits I used to represent the numbers go from 0 to 7, which is i = 0; i < (1 << n), since 1 << n = 2 ^ 3 = 8.Hope this clears some questions. :)
•  » » It stores the sum, in a specified position of the 'result array', of a particular subset. For better explanation look above. Thanks. :-)
 » Helped me a lot.Thanks...
 » Thanks a lot!!!It helped me a lot bhai...
 » What is the need for the down-votes? The writer has mentioned it is copied from a book, but it has still helped people looking for a quick answer in the end.
 » for ( x = y; x > 0; x = ( y & (x-1) ) )generates all subsets of bitmask y.How does this iteration works? Any intuitive explaination
•  » » 2 months ago, # ^ | ← Rev. 3 →   Well I couldn't understand how this generates subsets of y. But here is an alternative of it which I can explain. int b=0; do { cout<
•  » » » can't see the explaination you have provided, i just see a broken image link
•  » » » » I've edited my comment. Now you can see the link
 » dont know why downvotes for such a good editorial thanks @suraj021 for this excellent blog
 » Thank you Suraj for a clear and understandable post:D
 » Hey guys!I am a big noob at programming and I believe this book can surely help you at the big ACM ICPCs.Competitive Programming 3 by Steven Halim.
•  » » I hope you are not Steven Halim.
•  » » » Nope I am far from him. :)