Foundations of AI: From Machine Learning to Deep Learning

Define Artificial Intelligence and explain how Machine Learning fits within AI, using everyday examples (voice assistants, recommendations) and the analogy of teaching a child through examples versus giving step-by-step instructions — then answer: 'Is a calculator AI? Why or why not?'

Distinguish between supervised, unsupervised, and reinforcement learning using real-world analogies (e.g., studying with an answer key vs. sorting laundry vs. learning a video game) and identify which type applies to a given problem scenario.

Describe how a simple ML model learns from data: the intuition behind training, predictions, errors, and improving over time — using the analogy of adjusting your aim when throwing darts at a target.

Outline the initial stages of an ML project (problem definition, data collection, and data preparation), explaining the purpose of each stage using a concrete example such as predicting student exam scores.

Outline the core modeling and deployment stages of an ML project (model training, evaluation, and deployment) using the student score prediction example, then answer: 'How would we measure if our prediction model is any good?'

Define Deep Learning as a subfield of Machine Learning and explain the basic structure of a neural network (input/hidden/output layers, neurons, weights) using the analogy of an assembly line where each station refines the product further.

Contrast Machine Learning and Deep Learning by comparing their typical use cases, data requirements, and the role of feature engineering, with clear examples of when to choose one over the other.

Analyze case studies of AI applications (e.g., spam filters, medical image diagnosis), mapping them to specific ML/DL techniques, and outline criteria for selecting a viable AI project in a Q&A format.