Artificial Intelligence: Master Study Guide
Core Concepts • Machine Learning • Deep Learning • Generative AI • Applications • Ethics • MCQs • October 22, 2026
1. What Is Artificial Intelligence?
Artificial Intelligence (AI) is a field of computing focused on systems that perform tasks associated with human intelligence,
such as recognizing patterns, processing language, making predictions, planning, and generating content. AI systems differ
in design: some follow fixed rules, while others learn patterns from data.
Key terms: model, data, training, inference, feature, label, prediction, evaluation, automation, and human oversight.
2. Main Areas of AI
Machine Learning (ML): Algorithms learn patterns from examples to make predictions or decisions.
Deep Learning: A branch of ML using multi-layer neural networks to learn complex patterns from text, images, audio, and
other data.
Natural Language Processing: Methods for processing and generating human language.
Computer Vision: Methods for interpreting images and video, including classification and object detection.
Generative AI: Models that generate content such as text, images, audio, video, or code based on learned patterns.
AI agents: Systems designed to pursue a defined goal through steps and, when authorized, use tools or external
information.
3. Machine Learning Workflow
Define the problem → collect suitable data → clean and prepare it → split data into training, validation, and test sets → train
→ evaluate → deploy → monitor.
Supervised learning: Learns from labeled examples, such as spam classification.
Unsupervised learning: Finds patterns in unlabeled data, such as customer clustering.
Reinforcement learning: Learns through actions and rewards in an environment.
Overfitting: A model fits training details too closely and performs poorly on new data. Underfitting: A model fails to learn
enough useful structure.
Data leakage: Information that should not be available during prediction improperly influences training or evaluation.
4. Common Algorithms and Metrics
Method Typical use
Linear / Logistic Regression Numeric prediction / classification
Decision Tree / Random Forest Rule-based and ensemble predictions
K-Means Grouping similar observations
Neural Networks Learning complex patterns
Precision, Recall, F1 Classification evaluation
MAE / RMSE Numeric prediction error
5. Neural Networks and Deep Learning
Neural networks contain connected computational units arranged in layers. During training, parameters are adjusted to
reduce a loss function. Backpropagation calculates gradients, and an optimizer updates parameters. Results depend on
data quality, architecture, training, and evaluation.
6. Generative AI and Large Language Models
Core Concepts • Machine Learning • Deep Learning • Generative AI • Applications • Ethics • MCQs • October 22, 2026
1. What Is Artificial Intelligence?
Artificial Intelligence (AI) is a field of computing focused on systems that perform tasks associated with human intelligence,
such as recognizing patterns, processing language, making predictions, planning, and generating content. AI systems differ
in design: some follow fixed rules, while others learn patterns from data.
Key terms: model, data, training, inference, feature, label, prediction, evaluation, automation, and human oversight.
2. Main Areas of AI
Machine Learning (ML): Algorithms learn patterns from examples to make predictions or decisions.
Deep Learning: A branch of ML using multi-layer neural networks to learn complex patterns from text, images, audio, and
other data.
Natural Language Processing: Methods for processing and generating human language.
Computer Vision: Methods for interpreting images and video, including classification and object detection.
Generative AI: Models that generate content such as text, images, audio, video, or code based on learned patterns.
AI agents: Systems designed to pursue a defined goal through steps and, when authorized, use tools or external
information.
3. Machine Learning Workflow
Define the problem → collect suitable data → clean and prepare it → split data into training, validation, and test sets → train
→ evaluate → deploy → monitor.
Supervised learning: Learns from labeled examples, such as spam classification.
Unsupervised learning: Finds patterns in unlabeled data, such as customer clustering.
Reinforcement learning: Learns through actions and rewards in an environment.
Overfitting: A model fits training details too closely and performs poorly on new data. Underfitting: A model fails to learn
enough useful structure.
Data leakage: Information that should not be available during prediction improperly influences training or evaluation.
4. Common Algorithms and Metrics
Method Typical use
Linear / Logistic Regression Numeric prediction / classification
Decision Tree / Random Forest Rule-based and ensemble predictions
K-Means Grouping similar observations
Neural Networks Learning complex patterns
Precision, Recall, F1 Classification evaluation
MAE / RMSE Numeric prediction error
5. Neural Networks and Deep Learning
Neural networks contain connected computational units arranged in layers. During training, parameters are adjusted to
reduce a loss function. Backpropagation calculates gradients, and an optimizer updates parameters. Results depend on
data quality, architecture, training, and evaluation.
6. Generative AI and Large Language Models
