Tic-Tac-Toe Game

Dr. Ashish Kumar

Associate Professor-CSE
Manipal University Jaipur
T
 Program 2
• Board : A nine-element vector representing the board. We store 2 (indicating Blank), 3 (indicating X), or
5 (indicating O). Turn: An integer indicating which move of the game about to be played. The 1st move
will be indicated by 1, last by 9.
The Algorithm : The main algorithm uses three functions.
• Make_2 : returns 5 if the center square of the board is blank i.e., if board[5]=2. Otherwise, this function returns any non-
corner square (2, 4, 6 or 8).
• Posswin(p) : Returns 0 if player p can’t win on his next move; otherwise, it returns the number of the square that
constitutes a winning move. This function will enable the program both to win and to block opponents win. This function
operates by checking each of the rows, columns, and diagonals. By multiplying the values of each square together for an
entire row (or column or diagonal), the possibility of a win can be checked. If the product is 18 (3 x 3 x 2), then X can win.
If the product is 50 (5 x 5 x 2), then O can win. If a winning row (column or diagonal) is found, the blank square in it can
be determined and the number of that square is returned by this function.
• Go(n) : makes a move in square n. This procedure sets board [n] to 3 if Turn is odd, or 5 if Turn is even. It also increments
turn by one.
• The algorithm has a built-in strategy for each move. It makes the odd numbered move if it plays X, the even-numbered
move if it plays O.
Program 2
Program 2
 Program 2 : Pseudo Code
• Turn = 1 Go(1) (upper left corner).
• Turn = 2 If Board[5] is blank, Go(5), else Go(1).
• Turn = 3 If Board[9] is blank, Go(9), else Go(3).
• Turn = 4 If Posswin(X) is not 0, then Go(Posswin(X)) i.e. [ block opponent’s win], else Go(Make_2).
• Turn = 5 If Posswin(X) is not 0 then Go(Posswin(X)) [i.e. win], else if Posswin(O) is not 0, then Go(Posswin(O))
[i.e. block win], else if Board[7] is blank, then Go(7), else Go(3). [to explore other possibility if there be any ].
• Turn = 6 If Posswin(O) is not 0 then Go(Posswin(O)), else if Posswin(X) is not 0, then Go(Posswin(X)), else
Go(Make2).
• Turn = 7 If Posswin(X) is not 0 then Go(Posswin(X)), else if Posswin(X) is not 0, then Go(Posswin(O)) else go
anywhere that is blank.
• Turn = 8 If Posswin(O) is not 0 then Go(Posswin(O)), else if Posswin(X) is not 0, then Go(Posswin(X)), else go
anywhere that is blank.
• Turn = 9 Same as Turn=7.
 Iteration
Turn 1 (Agent 1's turn): Agent 1 (O) checks lines: 1, 2, 3, Turn 2 (Agent 2's turn): Agent 2 (X) checks lines: 1, 2, 3, 4,
4, 5, 6, 7, 8, 9, 1-5-9, and 3-5-7. No potential winning lines 5, 6, 7, 8, 9, 1-5-9, and 3-5-7. No potential winning lines for
for O are found. Agent 1 chooses the center position, 5. X are found. Agent 2 chooses position 8.

Turn 3 (Agent 1's turn): Agent 1 (O) checks lines: 1, 2, 3, 4, Turn 4 (Agent 2's turn): Agent 2 (X) checks lines: 1, 2, 3, 4, 5,
5, 6, 7, 8, 9, 1-5-9, and 3-5-7. No potential winning lines for 6, 7, 8, 9, 1-5-9, and 3-5-7. No potential winning lines for X are
O are found. Agent 1 chooses position 1. found. Agent 2 chooses position 3.
 Iteration
Turn 5 (Agent 1's turn): Agent 1 (O) checks lines: 1, 2, 3, 4, 5, 6,
7, 8, 9, 1-5-9, and 3-5-7. Potential winning line found for O: 1, 5,
9. Agent 1 chooses position 9.

Agent 1 (O) Wins!

Comments
• Not as efficient as first one in terms of time.
• Several conditions are checked before each move.
• It is memory efficient.
• Easier to understand & complete strategy has been
• determined in advance
• Still can not generalize to 3-D.
Implementations

With Ratings

With Magic Square

With Passwin

Learn Magic Square creation

Heuristic Search

Dr. Ashish Kumar

Associate Professor-CSE
Manipal University Jaipur
 Heuristics
• Psychologists Daniel Kahneman and Amos Tversky have developed the study of Heuristics in human
decision-making in the 1970s and 1980s.
• However, this concept was first introduced by the Nobel Laureate Herbert A. Simon, whose primary
object of research was problem-solving.
• Heuristic - a “rule of thumb” used to help guide search
• often, something learned experientially and recalled when needed
• Heuristic Function - function applied to a state in a search space to indicate a likelihood of success if
that state is selected
• Heuristic functions provide estimates of how close a state is to the goal. They guide search algorithms by
evaluating potential states.
• heuristic search methods are known as “weak methods” because of their generality and because they do not apply
a great deal of knowledge
• the methods themselves are not domain or problem specific, only the heuristic function is problem specific
 Heuristics
• Heuristic search is an AI search technique that employs heuristic for its moves.
• Heuristic is a rule of thumb that probably leads to a solution.
• Heuristics play a major role in search strategies because of exponential nature of the most problems.
Heuristics help to reduce the number of alternatives from an exponential number to a polynomial
number.
• In Artificial Intelligence, heuristic search has a general meaning, and a more specialized technical
meaning.
• In a general sense, the term heuristic is used for any advice that is often effective but is not guaranteed
to work in every case.
• Within the heuristic search architecture, however, the term heuristic usually refers to the special case of
a heuristic evaluation function.
 Heuristics
• Example: The Traveling Salesman Problem

• For a low number of cities, this question could be reasonably brute-forced. However, as the number of cities increases, it
becomes increasingly difficult to come to a solution.
• The nearest-neighbor (NN) heuristic solves this problem nicely: the computer always picks the nearest unvisited city next
on the path. NN does not always provide the best solution, but it is close enough to the best solution that the difference is
often negligible for the purpose of answering the TSP. By using this heuristic, the Big-O complexity of TSP can be reduced
from O(n!) to O(n^2).
 Applying Heuristics to Your Algorithms
• To apply heuristics to your algorithms, you need to know the solution or goal you’re looking for ahead of
time. If you know your end goal, you can specify rules that can help you achieve it.
• If the algorithm is being designed to find out how many moves a knight can make on a square, 8x8
chessboard while visiting every square, it’s possible to create a heuristic that causes the knight to always
choose the path with the most available moves afterward.
• However, because we’re trying to create a specific path, it may be better to create a heuristic that causes
the knight to choose the path with the fewest available moves afterward.
• Since the available decisions are much narrower, so too are the available solutions, and so they are
found more quickly.
Generate and Test Search

Dr. Ashish Kumar

Associate Professor-CSE
Manipal University Jaipur
 Generate and Test Search
• Generate and Test Search is a heuristic search technique based on Depth First Search with Backtracking
which guarantees to find a solution if done systematically and there exists a solution.
Algorithm
• Generate a possible solution.
• Test to see if this is actually a solution.
• Quit if a solution has been found.
• Otherwise, return to step 1.
 Generate and Test Search
• The evaluation is carried out by the heuristic function as all the solutions are generated systematically
in generate and test algorithm.
• But if there are some paths which are most unlikely to lead us to result then they are not considered.
• The heuristic does this by ranking all the alternatives and is often effective in doing so.
• But there is a technique to improve in complex cases as well by combining generate and test search with
other techniques so as to reduce the search space.
• It is also known as British Museum Search Algorithm as it’s like looking for an exhibit at random or
finding an object in the British Museum by wandering randomly.
• It is also known as brute-force search or exhaustive search.
 Key Characteristics of Generate and Test
• Dual Phases: Generation and testing are distinct but interconnected phases.

• Iterative Nature: Iteration allows for learning from previous attempts. Solutions are refined through
repeated iterations.

• Evaluation Criteria: Solutions are evaluated based on predefined criteria.

 Generation Phase
• The Generate phase is where the creative process takes place. Potential solutions are generated based on
the problem's context and requirements.

• Methods of Solution Generation

1. Systematic Enumeration: Solutions are generated systematically by exploring all possible
combinations.
2. Random Generation: Solutions are generated randomly, allowing for a diverse range of possibilities.

Example: Generating Solutions

• For a puzzle-solving problem, systematic enumeration might involve considering all possible sequences
of moves.
• In contrast, random generation could create various initial configurations for a puzzle.
 Testing Phase
• The Test phase is a critical step in the Generate and Test approach. Its purpose is to evaluate the
generated solutions for their validity and suitability.

Criteria for Testing Solutions

• Solutions are tested against specific criteria or constraints.
• These criteria could involve performance measures, feasibility, or adherence to predefined rules.

Example: Testing Solutions

• In a scheduling problem, solutions could be tested for meeting time constraints and resource
availability.
• For a design problem, solutions might be tested for structural integrity and cost-effectiveness.
 Pros & Cons of Generate and Test Search
Advantages –
• Simplicity and Flexibility: The concept of generating solutions and testing them is straightforward. It can be
adapted to a wide range of problems across various domains.
• Exploration of Solution Space: The approach explores diverse solution possibilities, even unconventional ones.
This exploration can lead to innovative and unexpected solutions.
• Applicability: Well-suited for complex problems with no direct formulaic solution. Effective for problems where
solution approaches are not well-defined.
Limitations –
• Inefficient for Large Solution Spaces: In scenarios with a vast solution space, exhaustive generation and testing
can be time-consuming and resource-intensive.
• Defining Testing Criteria: Establishing precise testing criteria can be complex, especially for subjective or
multidimensional problems.
• Limited Search Scope: The approach may miss optimal solutions if the generation process doesn't explore all
potential areas of the solution space.
Hill Climbing Search

Dr. Ashish Kumar

Associate Professor-CSE
Manipal University Jaipur
 Hill Climbing Search
• This algorithm is considered to be one of the simplest procedures for implementing heuristic search.
• Hill climbing is a simple, yet effective local search algorithm used in artificial intelligence and
optimization problems in mathematics.
• It is inspired by the idea of climbing a hill to reach the highest point, where the goal is to find the best
possible solution within a given problem space.
• Hill climbing search algorithm is simply a loop that continuously moves in the direction of increasing
value. It stops when it reaches a “peak” where no neighbour has higher value.
• This heuristic combines the advantages of both depth first and breadth first searches into a single
method.
• Hill climbing uses knowledge about the local terrain, providing a very useful and effective heuristic for
eliminating much of the unproductive search space. It is a branch by a local evaluation function.
• Local Search Technique + Greedy Approach + No Backtrack
 Hill Climbing Algorithm
Basic Idea:
• Hill climbing is a heuristic search algorithm that starts from an initial solution and iteratively moves to neighboring
solutions that are better in terms of the defined objective function. The algorithm keeps moving uphill (towards better
solutions) until it reaches a point where no better neighbor exists. At that point, it considers the current solution as a
local optimum.
Algorithm Steps:
• Initialization: Start from an initial solution within the problem space. This can be a randomly generated solution or a
predefined one.
• Evaluation: Evaluate the current solution using an objective function. The objective function defines the quality of the
solution and is used to determine whether a neighbor solution is better or worse.
• Neighbor Generation: Generate neighboring solutions by making small modifications to the current solution. These
modifications can include small changes or swaps, depending on the problem domain.
• Selection: Among the generated neighbors, select the one with the highest improvement according to the objective
function. If no neighbor improves the solution, the algorithm terminates as it has reached a local optimum.
• Iteration: Set the selected neighbor as the new current solution and repeat steps 2 to 4 until no further improvement is
possible.
Flow Chart of Hill Climbing Algorithm
 Types of Hill Climbing Algorithm
1. Simple Hill Climbing: This approach checks all neighbours one by one to see if any neighbour can provide a better
solution to the problem. The first encountered neighbour which provides a more optimal value is chosen as the next
current value. This search focus only on his previous and next step.
2. Steepest Ascent Hill Climbing: This approach builds up on the former and checks all the neighbours for solutions.
The neighbour which provides the best answer i.e., the steepest hill is considered to be the next current value. It
considers all the successive nodes, compares them, and choose the node which is closest to the solution.
3. Stochastic Hill Climbing: This approach does not focus on all the nodes. It selects a neighbour at random and
checks it to see if it provides a better solution. Then, it is compared with the current state and depending on the
differences in the values, it decides whether to examine a different neighbour or continue with the one that was just
evaluated.
4. First-Choice Hill Climbing (Random Restart Hill Climbing): This approach attempts to overcome local optima
(solutions that are better than their immediate neighbors but not globally optimal) by randomly selecting a number
of new starting points and performing hill climbing from each. This increases the chances of finding the global
optimum by exploring different areas of the search space.
 Types of Hill Climbing Algorithm
5. Random Mutation Hill Climbing: In this variant, the current solution is randomly mutated multiple times to
generate new candidate solutions. The best among these candidates is chosen as the next current solution. This
approach introduces randomness and diversity in the search process.
6. Simulated Annealing: Simulated annealing is a probabilistic hill climbing technique. It allows for accepting
solutions that are worse than the current one with a certain probability, which decreases over time. This probabilistic
acceptance allows the algorithm to escape local optima and explore the search space more effectively.
7. Hill Climbing with Restart: This technique involves periodically restarting the hill climbing algorithm from a
random or pre-defined starting point. The idea is to escape local optima and explore different parts of the search
space by restarting the process.
8. Parallel Hill Climbing: In this approach, multiple instances of the hill climbing algorithm run concurrently, each
starting from a different initial solution. This can speed up the search process by exploring multiple areas of the
search space simultaneously.
9. Memory-Based Hill Climbing: This variant maintains a memory of visited solutions and their evaluations. The
algorithm avoids revisiting previously explored solutions and focuses on unexplored or promising regions of the
search space.
 Problems with Hill Climbing Algorithm
1. Local Maximum is a state which is optimally better than
all its neighbouring states. So, if the hill climbing
algorithm arrives at this point, this value will be
returned. Whereas Global Maximum is the best state
possible in the landscape. This is the highest point in
the state space landscape.
• To overcome the local maximum problem: Utilize the
backtracking technique. Maintain a list of visited states.
If the search reaches an undesirable state, it can
backtrack to the previous configuration and explore a
new path.
 Problems with Hill Climbing Algorithm
2. Flat Local Maximum/Plateau is an area where all neighbouring states are at the same elevation i.e. each
solution is as optimal as the others. So, any of these points could be chosen.
• Shoulder is a plateau region which ends with an uphill edge. In this situation, the plateau area acts like a
flat region, where all points are equally suitable solutions. But we know that although the plateau ends
with an uphill edge, the algorithm may not continue its search until it reaches the edge and prematurely
return the shoulder region point as the solution, instead of continuing to look for higher points.
• To overcome plateaus: Make a big jump. Randomly select a state far away from the current state. Chances
are that we will land in a non-plateau region.

3. Ridge: Any point on a ridge can look like a peak because movement in all possible directions is
downward. Hence the algorithm stops when it reaches this state.
• To overcome Ridge: In this kind of obstacle, use two or more rules before testing. It implies moving in
several directions at once.
 Pros & Cons of Hill Climbing Algorithm
Advantages –
• Hill Climbing is a simple and intuitive algorithm that is easy to understand and implement.
• It can be used in a wide variety of optimization problems, including those with a large search space and complex
constraints.
• Hill Climbing is often very efficient in finding local optima, making it a good choice for problems where a good solution
is needed quickly.
• The algorithm can be easily modified and extended to include additional heuristics or constraints.
Limitations –
• Hill Climbing can get stuck in local optima, meaning that it may not find the global optimum of the problem.
• The algorithm is sensitive to the choice of initial solution, and a poor initial solution may result in a poor final solution.
• Hill Climbing does not explore the search space very thoroughly, which can limit its ability to find better solutions.
• It may be less effective than other optimization algorithms, such as genetic algorithms or simulated annealing, for
certain types of problems.
state-space landscape
objective
global maximum
function
cost
global minimum
function
current
state
neighbors
Hill
Climbing
Example – House & Hospital Distance
• The "House and Hospital" problem is a classic optimization problem that involves
finding the optimal assignment of houses to hospitals based on certain criteria. The
problem statement typically goes something like this:
Problem Statement:
• We have a set of houses and a set of hospitals. And our goal might be, in a world
that's formatted as the grid. Our objective in this case is to try and minimize the
distance of any of the houses from a hospital.
• There are a number of ways we could calculate that distance, but one way is the
Manhattan distance, this idea of how many rows and columns would you have to
move inside of this grid layout in order to get to a hospital.
Example – House & Hospital Distance
Cost: 17
Cost: 17
Cost: 17
Cost: 15
Cost: 13
 Cost: 11

This is optimal as per local maxima of the hill climbing algo. But What about the Global maxima?
 Cost: 9

This is optimal as per global maxima but hill climbing will not give this solution.
 Conclusion
• Remember that while hill climbing algorithms are relatively simple and intuitive, they may struggle with
complex optimization problems that have many local optima or discontinuous search spaces.

• To address the limitations of hill climbing algorithm, researchers have developed various extensions
and hybrid approaches that combine hill climbing with other optimization methods, such as genetic
algorithms, simulated annealing, or particle swarm optimization.

• These combinations aim to leverage the strengths of different techniques while mitigating their
weaknesses.

• It's important to convey that while hill climbing algorithms provide a foundational understanding of
optimization, they are just one piece of the broader landscape of optimization techniques.
 Simulated Annealing
• A variation of hill climbing in which, at the beginning of the process, some downhill moves may be
made.
• To do enough exploration of the whole space early on, so that the final solution is relatively insensitive to
the starting state.
• Lowering the chances of getting caught at a local maximum, or plateau, or a ridge.

Temperature Reduction Rules

 Simulated Annealing
Physical Annealing Algorithms:
• Physical substances are melted and then function SIMULATED-ANNEALING(problem, max):
gradually cooled until some solid state is current = initial state of problem
reached. for t = 1 to max:
• The goal is to produce a minimal-energy T = TEMPERATURE(t)
state. neighbor = random neighbor of current
Annealing schedule ΔE = how much better neighbor is than current
• If the temperature is lowered sufficiently if ΔE > 0:
slowly, then the goal will be attained. current = neighbor
• Nevertheless, there is some probability for a with probability eΔE/T set current = neighbor
transition to a higher energy state: e-E/kT. return current
• In Simple Hill Climb, It will move towards right upper
state and stuck at lower maxima.
• Whereas in Simulated Annealing, It will move towards
left lower state and then reach global maxima.
“Thank you for being such an
engaged audience during my
presentation.”
- Dr. Ashish Kumar