Invasion Game - Artificial Intelligence
Budget: ₹1,500 – ₹12,500 INR
Alice and Bob are playing a game with the following setting and rules:
There is an N*M grid which represents a kingdom which is being invaded by the players. Each
cell of the grid is either 1 or 0 where 1 denotes that there is land on this cell, while 0 denotes
water. In one turn, a player can capture a cell which has land, given that the cell has not been
captured before. If this cell shares a side with other uncaptured cells of land, the player can take
these in the same turn as well. A cell once captured by a player cannot be captured again by
another player. The game ends when all cells having land have been captured. The player who
has the most number of land cells wins.
If Alice plays the first turn, and both players play optimally, find the maximum number of cells
of land that Bob can capture.
Example:
Input:
4 4 #N,M
1001
0110
0110
1001
Output: 2
Note: The code will be tested against multiple inputs for marking. Hence, write a clean function
to read the input format and display the output. Though Adverserial search is the best fit for this
problem, instead design the next best suited uninformed search technique for this.
1. Explain the environment of the player agent in the game [20% weightage]
2. Explain the algorithm you will use for the solution and argue its correctness [20%
weightage]
3. Implement the solution using traversal/search to make the best choice for each player at
each stage [30% weightage]
4. Using pseudocode notation for your solution, outline the space and time complexity of
the solution with appropriate explanation [15% weightage]
5. Code’s correctness with other unseen test cases (during correction) [15% weightage]
Please note that marks will be deducted for not writing code which takes in the input in a
specific format (given in the adjoining notebook) properly and displays the required
output.
Submission guidelines:
● You are provided with the python notebook template which stipulates the structure of
code and documentation. Use well indented python code.
● The implementation code must be completely original and executable.
● Please keep your work (code, documentation) confidential. If your code is found to be
plagiarized, you will be penalized severely. Parties involved in the copy will be
considered equal partners and will be penalized severely
Topics
1. Uninformed search
2. Informed Search
3. Local search
justification of algorithms
word document: This should have justification towards the choices of
algorithm/data structures/heuristic/fitness function wherever appropriate written in your own words.
A screenshot of the output should be appended in the document and the code execution too.
There is an N*M grid which represents a kingdom which is being invaded by the players. Each
cell of the grid is either 1 or 0 where 1 denotes that there is land on this cell, while 0 denotes
water. In one turn, a player can capture a cell which has land, given that the cell has not been
captured before. If this cell shares a side with other uncaptured cells of land, the player can take
these in the same turn as well. A cell once captured by a player cannot be captured again by
another player. The game ends when all cells having land have been captured. The player who
has the most number of land cells wins.
If Alice plays the first turn, and both players play optimally, find the maximum number of cells
of land that Bob can capture.
Example:
Input:
4 4 #N,M
1001
0110
0110
1001
Output: 2
Note: The code will be tested against multiple inputs for marking. Hence, write a clean function
to read the input format and display the output. Though Adverserial search is the best fit for this
problem, instead design the next best suited uninformed search technique for this.
1. Explain the environment of the player agent in the game [20% weightage]
2. Explain the algorithm you will use for the solution and argue its correctness [20%
weightage]
3. Implement the solution using traversal/search to make the best choice for each player at
each stage [30% weightage]
4. Using pseudocode notation for your solution, outline the space and time complexity of
the solution with appropriate explanation [15% weightage]
5. Code’s correctness with other unseen test cases (during correction) [15% weightage]
Please note that marks will be deducted for not writing code which takes in the input in a
specific format (given in the adjoining notebook) properly and displays the required
output.
Submission guidelines:
● You are provided with the python notebook template which stipulates the structure of
code and documentation. Use well indented python code.
● The implementation code must be completely original and executable.
● Please keep your work (code, documentation) confidential. If your code is found to be
plagiarized, you will be penalized severely. Parties involved in the copy will be
considered equal partners and will be penalized severely
Topics
1. Uninformed search
2. Informed Search
3. Local search
justification of algorithms
word document: This should have justification towards the choices of
algorithm/data structures/heuristic/fitness function wherever appropriate written in your own words.
A screenshot of the output should be appended in the document and the code execution too.