The opening episode of the course. We build the mental map that everything else hangs on: AI is the umbrella goal, machine learning is the dominant method, deep learning is a branch of ML, and statistics and data science are overlapping siblings rather than nested layers.
Education segment covers:
- One-sentence map of AI, ML, deep learning, statistics, and data science, plus the operator-versus-user thesis of the whole arc.
- Definitions and history: the 1956 Dartmouth workshop, Arthur Samuel (1959), Tom Mitchell's 1997 textbook definition, the symbolic-to-statistical shift, AlexNet (2012), the Transformer (2017).
- Statistics vs ML through Leo Breiman's "Two Cultures" (2001): inference vs prediction.
- Worked examples: spam filtering (Paul Graham, "A Plan for Spam") and house-price prediction as both statistics and ML.
- The four paradigms: supervised, unsupervised, reinforcement, self-supervised.
- The ML lifecycle, the data-cleaning time myth, roles, and the portfolio-over-credential thesis.
- Market context: WEF Future of Jobs 2025, ML salary ranges.
News brief: