Artificial Intelligence (AI) Overview
1. Introduction to AI
- What is AI: Artificial Intelligence refers to the simulation of human intelligence in machines.
- Foundations of AI: Includes mathematics (logic, probability), neuroscience, computer engineering,
philosophy, and linguistics.
- History of AI: From early computing machines to modern deep learning models.
- AI - Past, Present and Future: Initial symbolic approaches, current machine learning models, and
future possibilities like general AI.
2. Intelligent Agents
- Environments: The surroundings within which an agent operates.
- Specifying the Task Environment: Defined by performance measure, environment, actuators, and
sensors (PEAS).
- Properties of Task Environments: Observable vs. partially observable, deterministic vs. stochastic,
episodic vs. sequential.
- Agent-Based Programs: Logic-based architecture to build intelligent agents.
- Structure of Agents: Consists of architecture and agent program.
- Types of Agents:
* Simple Reflex Agents: Act only on current percepts.
* Model-Based Reflex Agents: Use internal state to track aspects of the world.
* Goal-Based Agents: Act to achieve specified goals.
* Utility-Based Agents: Maximize a utility function.
3. Problem Solving by Searching
- Problem-Solving Agents: Formulate goals and search for solutions.
- Well-defined Problems: Have initial state, goal state, and a set of actions.
- Uninformed Search Strategies:
* Breadth-First Search, Uniform-Cost Search
* Depth-First Search, Depth-Limited Search
* Iterative Deepening Depth-First Search, Bidirectional Search
- Informed Search Strategies (Heuristic-Based):
* Greedy Best-First Search, A* Search, AO* Search
* Heuristic Functions: Estimate the cost from current to goal state
4. Knowledge Representation
- Knowledge-Based Agents: Use logical reasoning and knowledge.
- The Wumpus World: A simple example environment.
- Logic and Propositional Logic: Statements that are either true or false.
- Propositional Theorem Proving and Model Checking
- Agents Based on Propositional Logic
- First-Order Logic (FOL):
* Syntax and Semantics of FOL
* Using FOL: Express knowledge using variables, quantifiers
* Unification and Lifting: Matching expressions and generalizing
* Forward and Backward Chaining: Inference techniques
5. Learning
- Forms of Learning: Supervised, Unsupervised, Reinforcement
- Supervised Learning: Learning from labeled data
- Machine Learning:
* Decision Trees: Hierarchical model for decision making
* Regression and Classification with Linear Models
* Artificial Neural Networks: Model inspired by the brain
* Support Vector Machines: Finds optimal hyperplane for classification
6. Applications of AI
- Natural Language Processing: Understanding and generating human language
- Text Classification and Information Retrieval: Grouping and retrieving documents
- Speech Recognition: Converting speech into text
- Image Processing and Computer Vision: Understanding images and video
- Robotics: AI in machines capable of performing tasks
