<aside> ✅ Remember. This is intellectual property.

If you feel this resource is helpful and want to share it with your friends, please refer to @ghiffarin27 on Instagram or visit my website here

Thank you!

</aside>

1. AI (Artificial Intelligence)

• What it is: Any tech that mimics human thinking or decision-making.
• Core idea: Write rules or let machines learn so they can solve tasks like understanding speech, playing chess, or recommending videos.
• How it works:
  • Could be simple if-then rules (e.g., “if temperature > 30°C, turn on the fan”).
  • Or super complex, like systems that learn by themselves.
• Key traits:
  • Broad umbrella: covers everything from expert systems to chatbots.
  • Not all AI is “smart.” Some just follow predefined logic.
  • Encompasses rule-based, search algorithms, optimization, and learning methods.
• Example:
  • A chess engine that uses minimax search and heuristics to choose moves.
  • It isn’t learning from data—it follows a game tree and evaluation rules.

2. Machine Learning (ML)

• What it is: A subset of AI where machines learn patterns from data instead of relying on hardcoded rules.
• Core idea: Feed data in, adjust model parameters, and let the algorithm figure out relationships.
• How it works:
  • You give labeled data (for supervised) or unlabeled data (for unsupervised).
  • The model tunes itself to minimize error or find structure.
• Key traits:
  • Learns from examples. Doesn’t need every rule spelled out.
  • Improves performance as it sees more data (up to a point).
  • Can generalize to new, unseen data if trained well.
• Sub-branches:
  1. Traditional ML (classic algorithms with few layers)
  2. Deep Learning (neural networks with many layers)
• Example (ML overall):
  • Email spam filter. You show it thousands of labeled emails (“spam” or “not spam”), and it learns which words or patterns predict spam.

3. Traditional ML

• What it is: Classic ML algorithms that don’t use deep neural nets. Think of models built from straightforward math or trees.
• Core idea: Use statistical or geometric methods to draw boundaries or fit curves, based on features you engineer.