MODULE 1

Learning Outcomes:
At the end of this module, the students should be able to know and explain the
following:
1. Concept of Human and Artificial Intelligence.
2. Artificial/Computational Intelligence Paradigms.
3. Search, logic and Learning Algorithms.
1.1Human Intelligence
Human intelligence refers to the cognitive abilities of humans, including reasoning,
problem-solving, learning, memory, and perception. These abilities enable humans
to think, learn, and adapt to new situations.
1.2. Artificial Intelligence
Artificial intelligence (AI) is a branch of computer science that aims to create
intelligent machines that can think and behave like humans. AI systems are
designed to perform tasks that typically require human intelligence, such as
learning, reasoning, problem-solving, and perception.
1.2.1 Types of Artificial Intelligence
There are several types of AI, including:
 Weak AI or Narrow AI: Designed to perform a specific task, such as facial
recognition or language translation.
 General AI: Aims to create machines that can perform any intellectual task
that humans can.
 Super AI: A hypothetical type of AI that surpasses human intelligence in
every cognitive aspect.
2.1 Artificial Intelligence Paradigms
AI paradigms include: -
 Symbolic AI: Uses symbols and rules to represent knowledge and reason
about the world.
 Connectionist AI: Uses artificial neural networks to learn patterns in data.
 Evolutionary AI: Uses evolutionary algorithms to search for optimal
solutions.
1|Page
2.1.1 Applications of Artificial Intelligence
AI has numerous applications, including:
 Natural Language Processing: Enables computers to understand and
generate human language.
 Computer Vision: Enables computers to interpret and understand visual data.
 Robotics: Enables computers to control and interact with physical devices.
A. Symbolic AI Paradigm
Definition: Symbolic AI is a paradigm that uses symbols and rules to represent
knowledge and reason about the world.
Key Features:
(i) Uses formal languages to represent knowledge
(ii) Employs logical reasoning to deduce conclusions
(iii) Relies on explicit programming of knowledge and rules
Advantages:
 Can reason about abstract concepts.
 Can be used for expert systems and decision support systems.
 Can be integrated with other AI paradigms.
Disadvantages:
 Requires explicit programming of knowledge and rules.
 Can be brittle and inflexible.
 Can be difficult to scale to complex problems
B. Connectionist AI Paradigm
Definition: Connectionist AI is a paradigm that uses artificial neural networks to
learn patterns in data.
Key Features:
 Uses artificial neural networks to learn from data.
 Employs distributed representations to capture complex patterns.
 Relies on self-organization and learning from data
2|Page
Advantages:
 Can learn from large datasets.
 Can capture complex patterns and relationships.
 Can be used for image and speech recognition
Disadvantages:
 Requires large amounts of data to train.
 Can be difficult to interpret and understand.
 Can be prone to overfitting
C. Evolutionary AI Paradigm
Definition: Evolutionary AI is a paradigm that uses evolutionary algorithms to
search for optimal solutions.
Key Features:
(i) Uses evolutionary algorithms to search for optimal solutions
(ii) Employs principles of natural selection and genetics
(iii) Relies on iterative refinement and adaptation
Advantages:
 Can search for optimal solutions in complex spaces.
 Can adapt to changing environments and conditions.
 Can be used for optimization and planning
Disadvantages:
 Can be computationally expensive.
 Can be difficult to converge to optimal solutions.
 Can be prone to getting stuck in local optima
D. Hybrid AI Paradigm
Definition: Hybrid AI is a paradigm that combines multiple AI paradigms to
leverage their strengths.

Key Features:

3|Page
 (i) Combines symbolic, connectionist, and evolutionary AI paradigms
(ii) Employs hybrid architectures and algorithms
(iii) Relies on integration and fusion of multiple paradigms
Advantages:
 Can leverage the strengths of multiple paradigms.
 Can provide more robust and flexible solutions.
 Can be used for complex and dynamic problems
Disadvantages:
 Can be difficult to integrate multiple paradigms.
 Can be challenging to optimize and fine-tune hybrid systems.
 Can be prone to increased complexity and overhead
3.1 Search Algorithms
Search algorithms are used to find a solution to a problem by exploring a search
space.
3.1.1Types of Search Algorithms
A. Uninformed Search: Does not use any additional information about the
problem.
 Breadth-First Search (BFS).
 Depth-First Search (DFS).
 Uniform Cost Search (UCS)
B. Informed Search: Uses additional information about the problem to guide the
search.
 Best-First Search.
 A* Search.
 Greedy Search
Breadth-First Search (BFS)
1. Algorithm:
 Create a queue to hold nodes to be explored.
 Add the starting node to the queue.

4|Page
  While the queue is not empty.
 Dequeue a node.
 Explore the node's neighbors.
 Add the neighbors to the queue.
2. Time Complexity: O(b^d), where b is the branching factor and d is the depth of
the solution.
Depth-First Search (DFS)
3.1.2. Algorithm:
 Create a stack to hold nodes to be explored.
 Add the starting node to the stack.
 While the stack is not empty:
 Pop a node from the stack.
 Explore the node's neighbors.
 Add the neighbors to the stack.

3.2 Logic-Based Algorithms

Logic-based algorithms use logical reasoning to solve problems.
Propositional Logic
1. Definition: A logical system that uses propositions (statements that can be true
or false) to represent knowledge.
2. Operators: AND (∧), OR (∨), NOT (¬)
3.2.2 First-Order Logic
1. Definition: A logical system that uses predicates (functions that return true or
false) to represent knowledge.
2. Operators: ∀ (for all), ∃ (there exists), ∧ (AND), ∨ (OR), ¬ (NOT)
Resolution
1. Definition: A rule of inference that uses logical deduction to derive new
conclusions from existing knowledge.
2. Algorithm:

5|Page
  Start with a set of clauses (logical statements).
 Apply the resolution rule to derive new clauses.
 Repeat until a contradiction is found or a solution is derived.
3.3: Learning Algorithms
Learning algorithms are used to enable machines to learn from data.
3.3.2: Types of Learning
A. Supervised Learning: The machine is trained on labeled data to learn a
mapping between inputs and outputs.
B. Unsupervised Learning: The machine is trained on unlabeled data to discover
patterns and relationships.
C. Reinforcement Learning: The machine learns through trial and error by
interacting with an environment.
A. Supervised Learning Algorithms
1. Linear Regression: A linear model that predicts a continuous output variable.
2. Decision Trees: A tree-based model that predicts a categorical output variable.
3. Neural Networks: A complex model that predicts a continuous or categorical
output variable.
B. Unsupervised Learning Algorithms
1. K-Means Clustering: A clustering algorithm that groups similar data points into
clusters.
2. Principal Component Analysis (PCA): A dimensionality reduction algorithm
that transforms high-dimensional data into lower-dimensional data.
C. Reinforcement Learning Algorithms
1. Q-Learning: A model-free algorithm that learns to predict the expected return of
an action in a given state.
2. Deep Q-Networks (DQN): A model-free algorithm that uses a neural network to
approximate the Q-function.

6|Page