Google Interview Puzzle [closed]

Question

Given n dice, each of 'a' sides and a sum b, return the number of ways in which the sum b can be obtained. How can you reduce the time complexity and space complexity? This was asked in a Google interview and I am unsure of the answer.

Answer 1

This is asking you to find the number of ways to write b as a sum of n positive integers. The answer is the number of compositions of b into n parts, which is (b-1 choose n-1).

Now if we take into account the constraint that the size of the parts is limited to a, the problem gets a little more interesting. I recommend using generating functions for this. The answer will be the coefficient of x^b in the product (x+x^2+...+x^a)^n. Why? Because for each die (the singular of dice), you have a number between 1 and a, and this is represented by the exponent of x. When you take one x^i from each of the n terms, this is equivalent to the number i coming up on that die. The sum of the exponents must be the sum you are after, namely b.

We can even simplify the problem a bit using the multinomial theorem which states:

(x + x^2 + ... + x^a)^n = sum_{k1+k2+...+ka=n} (n multichoose k1,k2,...,ka) x^{k1+2*k2+...+a*ka}

where all ki >= 0. So the answer is that the number of ways is

sum_{k1+k2+...+ka=n & k1+2*k2+...+a*ka=b} (n multichoose k1,k2,...,ka)

Answer 2

I would have an array hits[max + 1] which counts the number of possible combination for each value. max is n * a and of course hits[0] to hits[n - 1] will remain empty.

The dumb way would be to do n for-loop (one for each die) and register a hit in hits for the current sum of the dice.

The less dumb way is to use a bit of combinatorics, where I write out the number of combinations for each ordered shuffle:

there is 1 combination of 1111 (sum = 4)
there are 4 combination of 1112 (sum = 5)
there are 4 combination of 1113 (sum = 6)
...
there are 4 * 3 / 2 combinations of 1123 (sum = 7)
...
there are 4 * 3 * 2 combinations of 1234 (sum = 10)
...
there is 1 combination of aaaa (sum = n * a)

You need to spend much less times in the for-loops than the dumb solution.
You get many hits for each iteration instead of just one hit with the dumb method.

Those for-loops just move the (n - 1) partition separations over (1, 2, 3, 4, ..., a). The separations can be on the same spot (e.g. they are all between 1 and 2 for the case 1111), but you must not have a separation below 1 or above a.

Answer 3

Assuming a is large enough( to be specific a>=b-n), this boils down to

x1+x2+x3+...+xn=b

which is the typical problem of distributing 'b' candies amongst 'n' kids. if you want to avoid 0 faced dies it should be easy to see that

(y1+1)+(y2+1)...+(yn+1)=b
y1+y2+...+yn=b-n

so a general solution to z1+z2+...zk=n is C(n+k-1,k-1)

EDIT after reciving several downvotes:

Assuming we have limit on 'a' i.e. b-n>a, we can formulate it as a DP problem where

dp[k][j] is no. of ways to get a sum of j using dices 1 to k inclusive
dp[1][j] is 0 if j>a or j==0 else 1

then we can have evaluate following relation

from k = 2 to n
  from j = 1 to b
     from x = 1 to a
        dp[k][j] += dp[k-1][j-x] where x is from 1 to a at max and x<j

and answer should be dp[n][b] The storage if of order n*b and runtime O(n*b*a)