## UVA Problem 10696 – f91 Solution

UVA Problem 10696 – f91 Solution:

Solving Technique:

This is a simple DP problem (OR, so it looks but there’s a better way). It can be solved both iteratively and recursively with or without memoization. Still it will get accepted. But still the runtime is slow due huge input. So using a faster io may help it get accepted with a better rank.

For the matter of practice i will show all four methods i know of to solve this problem. Codes are given in this order,

1. dynamic programming solution.
2. Recursive solution without Memoization.
3. Recursive solution with Memoization.
4. Shortcut Trickery.
5. Maybe extend the shortcut Trickery with memoization or other technique by yourself ? try it !!

#### Solution explanation:

Here the formula is given, we just need rearrange it and get the recurrence relation,

If N ≤ 100, then f91(N) = f91(f91(N+11));
If N ≥ 101, then f91(N) = N-10.

So from the given statement generating recurrence relation will be,

```f(n) = { f(f(n+11)), if n <= 100
{ n - 10,     if n >= 101
```

Here base case is for any n greater than or equal to 101 is return n – 10. No more recursive call on that.
This can be simplified but i won’t go into that. The codes below will give a better idea.

Now, just create a function named f and use if else or ternary to implement the given relation and it’ll work.

#### Memoization:

Since we’ve found out the recurrence relation it won’t be hard to memoize the code. We can use a memo array to store the already calculated results. We need not change much from the previous recursive solution to memoize it.

Add an if condition in the beginning if the value is already calculated and stored in memo array just return. Other wise when returning just store the return values in memo.

That’s it.

#### Dynamic Programming ( bottom up method ):

Our recurrence relation will never change. Since there is no other function needed we just calculate and store values in memo array. I started from 1,000,000 because the problem statement said an integer may be at most 1,000,000. Just loop through that many times and store results in memo array.

##### Last Solution Trick:

It come to me when i was testing my own inputs. Anything less or equal 100 result in 91. Any greater value results in that value minus 10. Take a look at the code below to get the idea.

Important:  Be sure to add or print a new line after each output unless otherwise specified. The outputs should match exactly because sometimes even a space character causes the answer to be marked as wrong answer. Please compile with c++ compiler as some of my codes are in c and some in c++.

Input:

```500
91
1
5
190
100
101
0```

Output:

```490
91
91
91
180
91
91```

### Code Bottom Up (Iterative) Dynamic Programming:

```/**
* @author  Quickgrid ( Asif Ahmed )
* Problem: UVA 10696 - f91
* Type:    Bottom Up (Iterative) Dynamic Programming
*/

#include<cstdio>
#include<sstream>

using namespace std;

#define N 1000000

static int F[N];

int main(){
int n, i;

for(i = N; i >= 0; --i){
if(i >= 101)
F[i] = i - 10;
else
F[i] = F[F[i + 11]];
}

while(scanf("%d", &n) && n)
printf("f91(%d) = %d\n", n, F[n]);

return 0;
}
```

### Code Top Down (Recursive) Without Memoization:

```/**
* @author  Quickgrid ( Asif Ahmed )
* Problem: UVA 10696 - f91
* Type:    Top Down (Recursive) without Memoization
*/

#include<stdio.h>

int f(int n){
return (n >= 101) ? n - 10 : f(f(n + 11));
}

int main(){
int n;

while(scanf("%d",&n) && n)
printf("f91(%d) = %d\n", n, f(n));

return 0;
}
```

### Code Top Down (Recursive) with Memoization:

```/**
* @author  Quickgrid ( Asif Ahmed )
* Problem: UVA 10696 - f91
* Type:    Top Down (Recursive) with Memoization
*/

#include<stdio.h>

static unsigned F[1000000];

unsigned f(unsigned n){
if(F[n])
return F[n];

if(n >= 101)
return F[n] = n - 10;
else
return F[n] = f(f(n+11));
}

int main(){
unsigned n;

while(scanf("%u",&n) && n)
printf("f91(%u) = %u\n", n, f(n));

return 0;
}
```

### Shortcut Technique:

```/**
* @author  Quickgrid ( Asif Ahmed )
* Problem: UVA 10696 - f91
* Type:    shortcut technique
*/

#include<stdio.h>

int main(){
unsigned n;

while(scanf("%u", &n) && n){
if(n <= 100)
printf("f91(%d) = 91\n", n);
else
printf("f91(%d) = %d\n", n, n - 10);
}

return 0;
}
```

## UVA Problem 10405 – Longest Common Subsequence Solution

UVA Problem 10405 – Longest Common Subsequence Solution:

Solving Technique:

This one is a fun problem. For this problem we need to find Longest Common Sub Sequence length of two given strings. We can solve this using LCS Algorithm discussed in Introduction to Algorithms book. This algorithm is explained in Wikipedia and other programming language implementations can be found here.

I have provided two dynamic programming implementations below with one top down memoized ( *Slower ) and a bottom up ( *Faster ) solution.

Take the string inputs carefully there may be empty lines in between.

Overview:

Simple explanation for solution technique is we apply Dynamic Programming techniques. There are overlapping sub problems that we can combine to get original solutions.

Starting from the end of both strings. If both of the characters in string are same then we can reduce both string size by 1 length. So now if do the reverse meaning add 1 we get original solution.

In case both characters are not same keeping one string same we reduce the other one by 1 length and try to match the last characters again.

This way if two characters are  same we increase LCS length count. Among the sub problem we choose the one with longest length.

Important:  Be sure to add or print a new line after each output unless otherwise specified. The outputs should match exactly because sometimes even a space character causes the answer to be marked as wrong answer.

Input:

```bcacbcabbaccbab
bccabccbbabacbc

a1b2c3d4e
zz1yy2xx3ww4vv

abcdgh
aedfhr

abcdefghijklmnopqrstuvwxyz
a0b0c0d0e0f0g0h0i0j0k0l0m0n0o0p0q0r0s0t0u0v0w0x0y0z0

abcdefghijklmnzyxwvutsrqpo
opqrstuvwxyzabcdefghijklmn```

Output:

```11
4
3
26
14```

### Bottom Up Memoized Code:

```/**
* @author  Quickgrid ( Asif Ahmed )
* Problem: UVA 10405 Longest Common Subsequence ( LCS )
* Method:  Memorized Recursive / Top Down Solution
*/

#include<stdio.h>;
#include<string.h>;
#define SIZE 1024

static char x[SIZE], y[SIZE];
static int lcs[SIZE][SIZE];

int maxval(int a, int b){
return a > b ? a : b;
}

/*
* If the value is not -1 then it means the value of that sub problem is
* already calculated. Just return that calculated value
* If both of the characters in string are same then we can reduce both string size by 1 length and calculate rest
* Else among sub problems by reducing one string and keeping the other one same find the one with the max length
*/
int LCS(int i, int j){
if(lcs[i][j] != -1)
return lcs[i][j];

if(x[i-1] == y[j-1])
lcs[i][j] = LCS(i-1, j-1) + 1;
else
lcs[i][j] = maxval( LCS(i-1, j), LCS(i, j-1) );

return lcs[i][j];
}

int main(){
register unsigned i, j;
while(gets(x) && gets(y)){

int xlen = strlen(x);
int ylen = strlen(y);

/*
* Set -1 to all positions to indicate there are no calculated value
*/
for(i = 1; i <= xlen; ++i)
for(j = 1; j <= ylen; ++j)
lcs[i][j] = -1;

printf("%d\n", LCS(xlen,ylen) );

}
return 0;
}
```

### Top Down DP Code:

```/**
* @author  Quickgrid ( Asif Ahmed )
* Problem: UVA 10405 Longest Common Subsequence ( LCS )
* Method:  Top Down Dynamic Programming Solution
*/

#include<stdio.h>
#include<string.h>
#define SIZE 1024

static char x[SIZE], y[SIZE];
static int lcs[SIZE][SIZE];

int maxval(int a, int b){
return a > b ? a : b;
}

int main(){
register int i, j;

while(gets(x) && gets(y)){

int xlen = strlen(x);
int ylen = strlen(y);

/*
* If both of the characters in string are same then we can reduce both string size by 1 length and calculate rest
* Else among sub problems by reducing one string and keeping the other one same find the one with the max length
*/
for(i = 1; i <= xlen; ++i){
for(j = 1; j <= ylen; ++j){
if(x[i-1] == y[j-1])
lcs[i][j] = lcs[i-1][j-1] + 1;
else
lcs[i][j] = maxval(lcs[i-1][j], lcs[i][j-1]);
}
}

/*
* The max length is at the bottom right corner of the table
*/
printf("%d\n", lcs[xlen][ylen] );

}
return 0;
}
```

## Fibonacci Memoization Top Down ( Recursive ) and Bottom Up ( Iterative )

Fibonacci Memoization Top Down ( Recursive ) and Bottom Up ( Iterative ) Dynamic Programming:

### Top Down Fibonacci Memoization Code:

```/**
* @author  Quickgrid ( Asif Ahmed )
*/

#include<stdio.h>

#define N 64
long long F[N];

/**
* Top Down Recursive Fibonacci Memoization
*/
long long topDownMemFib(long long n){
/**
* If its not 1 then it means there is already a calculated value
* So no more need to compute, just return that value
*/
if(F[n] != -1)
return F[n];

F[n] = topDownMemFib(n - 1) + topDownMemFib(n - 2);
return F[n];
}

int main(){
register unsigned i;

unsigned n;
printf("Enter a number:\n");
scanf("%u", &n);

/**
* Terminating condition values
*/
F[0] = 0;
F[1] = 1;

/**
* First set the whole array to -1, means there are no calculated value
*/
for(i = 2; i <= n; ++i)
F[i] = -1;

printf("%lld\n", topDownMemFib(n));

return 0;
}
```

### Bottom Up Iterative Fibonacci Implementation:

```/**
* @author  Quickgrid ( Asif Ahmed )
*/

#include<stdio.h>

/**
* Increase this value for numbers more than 64
*/
#define N 64
long long F[N];

/**
* Top Down Fibonacci Memoization
*/
long long bottomUpMemFib(long long n){
/**
* Terminating Condition
*/
F[0] = 0;
F[1] = 1;

register unsigned i;
for(i = 2; i <= n; ++i)
F[i] = F[i - 1] + F[i - 2];

return F[n];
}

int main(){
register unsigned i;

unsigned n;
printf("Enter a number:\n");
scanf("%u", &n);

printf("%lld\n", bottomUpMemFib(n));

return 0;
}
```