DP vs Recursion vs Greedy Choosing the Right Tool in DSA C - Visual Comparison

Concept Flow - DP vs Recursion vs Greedy Choosing the Right Tool

Start Problem

↓

Can problem be divided into smaller subproblems?

↓

Is there overlapping subproblems?

↓

Use DP (Memoization or Tabulation)

↓

Solve subproblems efficiently

↓

Combine results for final answer

↓

Done

This flow shows how to decide between recursion, dynamic programming, and greedy methods based on problem properties.

Execution Sample

DSA C

int fib(int n) {
  if (n <= 1) return n;
  return fib(n-1) + fib(n-2);
}

Simple recursion to compute Fibonacci number, recalculating subproblems multiple times.

Execution Table

Step	Operation	Nodes in Call Stack	Subproblem Computed	Result Returned	Visual State
1	Call fib(4)	fib(4)	None yet	Pending	Call stack: fib(4)
2	Call fib(3)	fib(4) -> fib(3)	None yet	Pending	Call stack: fib(4), fib(3)
3	Call fib(2)	fib(4) -> fib(3) -> fib(2)	None yet	Pending	Call stack: fib(4), fib(3), fib(2)
4	Call fib(1)	fib(4) -> fib(3) -> fib(2) -> fib(1)	fib(1) = 1	1	Return fib(1)=1, pop fib(1)
5	Call fib(0)	fib(4) -> fib(3) -> fib(2) -> fib(0)	fib(0) = 0	0	Return fib(0)=0, pop fib(0)
6	Return fib(2) = fib(1)+fib(0)	fib(4) -> fib(3)	fib(2) = 1	1	Return fib(2)=1, pop fib(2)
7	Call fib(1)	fib(4) -> fib(3) -> fib(1)	fib(1) = 1	1	Return fib(1)=1, pop fib(1)
8	Return fib(3) = fib(2)+fib(1)	fib(4)	fib(3) = 2	2	Return fib(3)=2, pop fib(3)
9	Call fib(2)	fib(4) -> fib(2)	fib(2) recomputed	Pending	Call stack: fib(4), fib(2)
10	Call fib(1)	fib(4) -> fib(2) -> fib(1)	fib(1) = 1	1	Return fib(1)=1, pop fib(1)
11	Call fib(0)	fib(4) -> fib(2) -> fib(0)	fib(0) = 0	0	Return fib(0)=0, pop fib(0)
12	Return fib(2) = fib(1)+fib(0)	fib(4)	fib(2) = 1	1	Return fib(2)=1, pop fib(2)
13	Return fib(4) = fib(3)+fib(2)	None	fib(4) = 3	3	Return fib(4)=3, call stack empty

💡 All recursive calls completed, final result fib(4) = 3 returned

Variable Tracker

Variable	Start	After Step 4	After Step 6	After Step 8	After Step 13
fib(4)	Not called	Pending	Pending	Pending	3
fib(3)	Not called	Pending	1	2	Returned
fib(2)	Not called	Pending	1	Recomputed	1
fib(1)	Not called	1	Returned	Returned	Returned
fib(0)	Not called	0	Returned	Returned	Returned

Key Moments - 3 Insights

Why does fib(2) get computed twice in simple recursion?

Because recursion calls fib(2) separately from fib(3) and fib(4) without saving results, as shown in steps 6 and 12 in the execution table.

When should we choose dynamic programming over simple recursion?

When subproblems overlap and get recomputed multiple times, like fib(2) here, DP saves results to avoid repeated work, improving efficiency.

Why is greedy not suitable for Fibonacci calculation?

Because Fibonacci depends on previous two values (optimal substructure) but no greedy choice can build the solution step-by-step without missing subproblems.

Visual Quiz - 3 Questions

Look at the execution table, what is the result returned by fib(3) at step 8?

A 1

B 2

C 3

D Pending

💡 Hint

Check the 'Result Returned' column at step 8 in the execution table.

At which step does the call stack become empty, indicating all recursion calls finished?

A Step 10

B Step 12

C Step 13

D Step 6

💡 Hint

Look at the 'Visual State' column for when the call stack is empty.

If we use dynamic programming, how would the number of fib(2) computations change compared to this recursion?

A It would reduce to one

B It would increase

C It would stay the same

D It would be zero

💡 Hint

Refer to variable_tracker and key_moments about overlapping subproblems.

Concept Snapshot

DP vs Recursion vs Greedy:
- Recursion solves by calling itself, may recompute subproblems.
- DP saves results to avoid repeated work (overlapping subproblems).
- Greedy picks best local choice, works if problem has optimal substructure.
- Choose DP if subproblems overlap.
- Choose Greedy if local choices lead to global optimum.
- Choose Recursion if problem is simple or no overlap.

Full Transcript

This concept helps decide when to use recursion, dynamic programming, or greedy algorithms. Recursion solves problems by calling itself but can repeat work. Dynamic programming improves recursion by saving results of subproblems to avoid recomputation, especially when subproblems overlap. Greedy algorithms pick the best choice at each step and work when this leads to the best overall solution. The example of Fibonacci numbers shows simple recursion calling fib(2) twice, which wastes time. Dynamic programming would compute fib(2) once and reuse it. Greedy is not suitable here because Fibonacci depends on previous two values, not a single local choice. The execution table traces each recursive call and return, showing the call stack growing and shrinking. Variable tracker shows how values change after key steps. Key moments clarify why repeated computation happens and when to choose each method. The visual quiz tests understanding of the execution steps and concepts. This helps beginners see how these approaches differ and when to pick the right tool for a problem.

Practice

1. Which approach is best suited when a problem has overlapping subproblems and optimal substructure?

easy

A. Dynamic Programming

B. Greedy Algorithm

C. Simple Recursion without memoization

D. Brute Force

Solution

1. Step 1: Understand problem characteristics

   Overlapping subproblems mean the same smaller problems repeat. Optimal substructure means the best solution can be built from best solutions of smaller problems.

2. Step 2: Match approach to characteristics

   Dynamic Programming stores results of subproblems to avoid repeated work, ideal for overlapping subproblems and optimal substructure.

3. Final Answer:

   Dynamic Programming -> Option A

4. Quick Check:

   Overlapping subproblems + optimal substructure = Dynamic Programming [OK]

Hint: Use DP when subproblems repeat and build optimal solutions [OK]

Common Mistakes:

• Choosing greedy when subproblems overlap
• Using simple recursion without memoization causing slow code
• Confusing optimal substructure with greedy choice property

2. Which of the following is the correct syntax to implement memoization in C for a recursive Fibonacci function?

easy

A. Use a global array to store results and check before recursive calls

B. Use a local variable inside the function to store results

C. Use a static variable without initialization inside the function

D. Use a pointer without allocating memory for memoization

Solution

1. Step 1: Understand memoization storage

   Memoization requires storing results outside the recursive calls to reuse them. A global or static array is suitable.

2. Step 2: Check options for correct usage

   Only a global array checked before recursion ensures stored results are reused properly. Local variables reset each call, static without init may cause errors, pointers need allocation.

3. Final Answer:

   Use a global array to store results and check before recursive calls -> Option A

4. Quick Check:

   Memoization needs persistent storage = global array [OK]

Hint: Memoize with global/static array checked before recursion [OK]

Common Mistakes:

• Using local variables that reset each call
• Not initializing static variables properly
• Using pointers without memory allocation

3. What will be the output of the following C code snippet implementing a greedy approach for coin change?

int coins[] = {25, 10, 5, 1};
int amount = 30;
int count = 0;
for (int i = 0; i < 4; i++) {
    while (amount >= coins[i]) {
        amount -= coins[i];
        count++;
    }
}
printf("%d\n", count);

medium

A. 3

B. 2

C. 4

D. 5

Solution

1. Step 1: Trace the greedy coin selection

   Start with 30 amount. Use 25 coin once (30-25=5), count=1. Then 10 coin can't be used (5<10). Use 5 coin once (5-5=0), count=2.

2. Step 2: Count total coins used

   Coins used: 25 (1), 5 (1), total count = 2 coins. The code increments count inside while loop, so count=2.

3. Step 3: Check for remaining coins

   Amount is now 0, so no more coins used.

4. Final Answer:

   2 -> Option B

5. Quick Check:

   Greedy coin count for 30 = 2 coins [OK]

Hint: Subtract largest coin repeatedly until amount zero [OK]

Common Mistakes:

• Counting coins incorrectly by missing loops
• Assuming greedy always gives minimum coins
• Confusing coin values and counts

4. Identify the error in this recursive function for factorial and suggest the fix:

int factorial(int n) {
    if (n == 0) return 0;
    return n * factorial(n - 1);
}

medium

A. Recursive call should be factorial(n + 1)

B. Function should return void

C. Base case should return 1 instead of 0

D. No error, code is correct

Solution

1. Step 1: Analyze base case correctness

   Factorial of 0 is 1, but code returns 0, which will cause all results to be zero.

2. Step 2: Suggest correct base case

   Change base case to return 1 when n == 0 to correctly end recursion and compute factorial.

3. Final Answer:

   Base case should return 1 instead of 0 -> Option C

4. Quick Check:

   Factorial(0) = 1 base case [OK]

Hint: Factorial base case returns 1, not 0 [OK]

Common Mistakes:

• Returning 0 in base case causing zero product
• Changing recursion direction incorrectly
• Ignoring base case leading to infinite recursion

5. You want to find the minimum number of coins to make a given amount using coins of denominations {1, 3, 4}. Which approach is best and why?

hard

A. Brute force, to try all combinations without optimization

B. Recursion without memoization, to explore all possibilities

C. Greedy, because always picking largest coin works

D. Dynamic Programming, to store and reuse subproblem results

Solution

1. Step 1: Understand problem requirements

   Minimum coins for amount with coins {1,3,4} needs checking all combinations to find optimal solution.

2. Step 2: Evaluate approaches

   Greedy fails here because picking largest coin first may not yield minimum coins. Recursion without memoization is slow. Dynamic Programming stores results of smaller amounts to build optimal solution efficiently.

3. Final Answer:

   Dynamic Programming, to store and reuse subproblem results -> Option D

4. Quick Check:

   Minimum coins with overlapping subproblems = DP [OK]

Hint: Use DP when greedy fails to find minimum coins [OK]

Common Mistakes:

• Assuming greedy always works for coin change
• Using recursion without memo causing slow code
• Ignoring overlapping subproblems in coin change