Introduction to Artificial Intelligence

Flashcards covering definitions, layers of AI, and the distinctions between AI, Machine Learning, Deep Learning, and Data Science based on the lecture notes.

Artificial Intelligence (AI)

The simulation of human intelligence in machines that are capable of performing tasks that normally require human intellect, such as reasoning, learning, and problem-solving.

AI as a Component

The concept that AI is often not an entire system itself (like Instagram), but rather specific components within a larger system used for recommendation, pattern recognition, or behavior modeling.

Automation

Software, such as a basic calculator, that follows predefined rules and executes deterministic mathematical functions without learning, adapting, or improving with experience.

Predictive Models

Tools used in AI systems like Google Maps to learn from past traffic patterns and predict future conditions to adapt dynamically.

Narrow AI

A classification of AI, such as ChatGPT, that is trained on massive data to perform specific tasks like predicting the next word probabilistically without human-like understanding.

Acting intelligently vs. actually being intelligent

The distinction that AI performs tasks that appear intelligent by imitating human behavior (Layer 1) without necessarily "thinking" like a human.

Rational agent

An AI system that perceives its environment, makes decisions, and chooses the best possible action to maximize success or achieve a specific goal (Layer 2).

Pattern Recognition

The basis of modern AI (Layer 3) where systems detect statistical patterns in images, text, audio, or user behavior rather than relying on rule-based logic.

Machine Learning (ML)

A method of achieving AI by allowing machines to learn patterns from data and adjust internal parameters to improve performance over time instead of being explicitly programmed.

Deep Learning (DL)

A specialized type of machine learning using artificial neural networks with many layers, inspired by the human brain, requiring large datasets and high computing power ($GPUs$).

Data Science

An interdisciplinary field involving statistics, data analysis, and programming focused on extracting insights and knowledge from data; it overlaps with but is not a subset of AI.

Traditional Programming vs. Machine Learning

Traditional programming uses explicit rules (If marks < $40$ \rightarrow Fail), while machine learning provides data (e.g., $10,000$ records) and lets the system discover the rules.