001-2024-1118_DLBDSEAIS01_Course_Book

Course Information

The course book serves as the core material for the study of Artificial Intelligence, delineated into distinct units with key concepts and learning objectives.
Additional Resources: Learning materials are also available on the learning platform, including self-check questions at the end of each section to ensure comprehension of concepts.

Introduction... (p. 7)
Basic Reading... (p. 9)
Further Reading... (p. 10)
Learning Objectives... (p. 11)
Unit 1 - History of AI: Covers historical developments, AI winters, expert systems, and notable advances in AI (pp. 12-28).
Unit 2 - Modern AI Systems: Differentiates between narrow and general AI, and application areas (pp. 29-36).
Unit 3 - Reinforcement Learning: Introduces principles of reinforcement learning, Markov Decision Processes, and Q-learning algorithm (pp. 37-46).
Unit 4 - Natural Language Processing: Discusses NLP, models and techniques, including vectorizing data (pp. 45-68).
Unit 5 - Computer Vision: Defines computer vision tasks, image representation, and feature detection (pp. 69-90).
Backmatter: References and index (pp. 91-102).

Gain familiarity with the field of artificial intelligence, its origins across disciplines like cognitive science and neuroscience.
Understand the historical shaping of AI and its paradigms.
Explore common tasks and applications within AI.
Master the fundamentals of reinforcement learning, natural language processing, and computer vision which are vital for AI agents to engage with their environment.

Early considerations of AI date back to Ancient Greece, but formal developments began in the 1950s.
Key figures included Aristotle, who formalized logical reasoning; Leonardo da Vinci, who imagined computational devices; and René Descartes, who connected rationality to mathematics.

The first AI winter from 1974 to 1980 occurred when machine translation expectations were not met.
The second AI winter from 1987 to 1993 was due to setbacks in developing expert systems.

Knowledge-based systems designed to make decisions using the expertise of a human specialist. Their components include:
- Knowledge Base: Stores domain-specific knowledge.
- Inference Engine: Draws conclusions from the knowledge base.
- User Interface: Allows interaction between the user and the system.

Differentiate between narrow AI (ANI) and general AI (AGI).
Identify major application areas for AI in various industries, including healthcare, finance, and automotive sectors.

Significant increases in AI adoption across various business functions, supported by substantial growth in related research.

Understand reinforcement learning, its principles, and utilization of Q-learning.

Reinforcement learning involves agents learning through trial and error in an environment to maximize rewards (positive or negative).
Markov Decision Processes (MDPs) provide a framework for modeling decision-making scenarios in reinforcement learning.

A model-free method where agents learn to evaluate the quality of actions taken in states to maximize the cumulative reward.

Define NLP and explore its applications in speech recognition, understanding, and generation.

Focus on techniques such as Bag-of-Words, statistical methods, and neural models for processing and generating language.
Important NLP tasks include topic identification, sentiment analysis, and machine translation.

Understand how images are represented, the tasks in computer vision, and how to correct distortions.

Computer vision aims to interpret visual data and includes tasks such as object detection and motion analysis.
Challenges include lighting variations, object identification, and the need for advanced feature detection algorithms.

This course provides a comprehensive overview of AI, its history, methodologies, and applications across various domains, including NLP and computer vision. The increasing computational power and adoption of AI reflect its significance in modern technology and society.