Scribd
Upload
8 views

AI A3 2023spring

The document outlines three problems related to artificial intelligence techniques: local search algorithms for solving the N-queens problem, adversarial search using minimax algorithm for tic-tac-toe, and constraint satisfaction problem of graph coloring. Students are asked to implement and compare algorithms to solve each problem.

Uploaded by

Sami Ahmed
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
8 views

AI A3 2023spring

The document outlines three problems related to artificial intelligence techniques: local search algorithms for solving the N-queens problem, adversarial search using minimax algorithm for tic-tac-toe, and constraint satisfaction problem of graph coloring. Students are asked to implement and compare algorithms to solve each problem.

Uploaded by

Sami Ahmed
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

National University

of Computer and Emerging Sciences, CFD Campus

AI 2002 - Artificial Intelligence (Spring 2023)


Assignment # 3

Topics Covered: Local search, adversarial Submission Deadline: April 03, 2023, by
search, constraint satisfaction problem 16.00 sharp

Submit your source files in a zipped folder on


the Google Classroom (of your respective
section)

Problem # 1: [Local Search Algorithms]


Solve the N-queens problem using the following two local search algorithms: (a) Hill Climbing,
and, (b) Simulated Annealing. Use the following to implement your algorithms.

● Program input: take no. of queens (N) as input from the user (any value between 5 to
10)
● Board configuration: on an NxN board, queens will be represented by the letter ‘Q’,
and empty cells will be represented as a ‘dash’ (-) character
● Start state: random placement of N queens on an NxN board (each in one column)
● Successor function: a successor or neighbor of a state can be generated by
randomly moving any one of the Queen inside that column
● Objective function to minimize: no. of queens attacking each other (commutative: Q1
attacking Q4 and vice verse is just one attack)
● Temperature function for Simulated Annealing: use the following link as a reference
to find different temperature functions
(https://mlrose.readthedocs.io/en/stable/source/decay.html), use ExpDecay in your
algorithm
● Program output: to compare the performance of the two algorithms, make sure that
both algorithms begin with the same start state. As both algorithms converge, print the
following as output.
● Success or Failure (i.e., goal state reached or not)?
● No. of moves made
● Which algorithm achieves the best minima?

Problem # 2: [Adversarial Search]


Implement the Tic-Tac-Toe game for two rational agents (Max and Min) who will play
adversaries to each other using the Minimax algorithm. If the two rational agents play against
each other in a zero-sum game, who will win actually?

Page 1
National University
of Computer and Emerging Sciences, CFD Campus

Given the following problem description, implement the following algorithms.


(a) Minimax algorithm (without pruning)
(b) Minimax with the Alpha-Beta pruning algorithm and show the level of the node which
was pruned (assume that the start state is at level 0)

● Random toss: declare who won the toss, Max or Min. The player who won the toss will
take the first move
● Players moves: represent Max move as an ‘X’ and Min move as an ‘O’
● Start state: the 3x3 board is empty initially
● Successor function: no. of empty cells that can be filled by a player
● Terminal states with utility values: Max wins (+1 utility), Min wins (-1 utility), Draw (0
utility)
● Program output:
○ Board configuration after each player’s move
○ Announce the winning player
○ No. of moves taken by each player as the game ends
○ Total no. of successor nodes generated by each player during search

Problem # 3: [Constraint Satisfaction Problem]


Given an undirected graph, we want to solve the graph coloring problem as described below.
● Variables: a, b, c, d, e, f, g
● Domain: R, G, B
● Constraint: any two neighbors cannot have the same color

Implement a solution to this problem using the following and use chronological order to select
the next variable to break a tie or if no heuristic is used.
(a) Solve the problem using AC-3.
(b) Solve the problem using DFS with backtracking and heuristics including MRV, DH, and,
LCV.

Page 2
National University
of Computer and Emerging Sciences, CFD Campus

Program output:
● Show the values of all variables after each assignment
● Total number of assignments made to solve the problem
● For (b), count the number of backtracking performed and the no. of calls made to each
heuristic

Page 3

You might also like