Here is the link to the problem: http://codeforces.net/contest/401/problem/D
So I was trying the same DP approach as most people did, but with a slight change in the state.
What most people used as their DP state: < remainder, mask_of_18_bits >
where 'mask_of_18_bits' is used to denote whether i-th digit in number 'n' is used or not.
What I used as DP state: < remainder, zeroes_left, mask_of_18_digits >
Please note that here the mask is entirely different.
where 'zeroes_left' = number of 0s left in the number 'n' to be used.
and 'mask_of_18_digits' is defined in this way 'aabbccddeeffgghhii'
aa = number of 9s left in 'n' to be used,
bb = number of 8s left in 'n' to be used,
....
....
ii = number of 1s left in 'n' to be used
Here is the link to my submission: http://codeforces.net/contest/401/submission/5990314
but with my approach, I used std::map and that gives TLE at test case 80.
Can someone please suggest a workaround using the same idea? I want to be able to define a state in terms of counts of the digits 0-9.
Take a look at my solution. I've used the same idea but encoded counts in binary bitmask.
It's amazing how you encoded all that into just one long long variable. Can you please brief me a bit about your encoding and decoding process?
EDIT: I have understood your method. Cool trick! Thanks for sharing!