Track quiz
Before the questions, here is each idea in one line.
| Idea | The line to keep |
|---|---|
| Both at once | Sort a machine by two facts. Did a person write the rule? Did a part learn from examples? |
| Training | The rule is learned from the pile. When the pile changes, the rule changes. |
| A decision tree | A decision tree asks one yes/no question at a time. Two things that end up together gave the same answers. |
| What it learned from | Ask what a machine was shown. Pictures and video usually mean computer vision. Words and speech usually mean NLP. Tables of numbers usually mean data science. |
| Following it back | The path is the questions it was asked, in order. |
A car park barrier reads plates with a learned part. It also checks a typed list. Which bin?
What makes a crate a labelled example?
Two more crates are added to the pile. They disagree with the rule. What can happen?
A program says each crate is full price or cut price. What is that called?
A program guesses the kilos of tomatoes a stall will sell tomorrow. What is that called?
A chain of three yes/no questions, each no leading to the next. How many ends?
You chose every question in a tree yourself. Is it a written rule or a learned one?
Eight things are in a box. One question takes out the two that say yes. How many are left?
A phone unlocks when it sees your face. Which kind of AI work is that?
A dairy guesses next month's orders from its past months. Which kind of work?
A tree sends a tray of paneer to the open shelf. The list says cold cupboard. Where is the mistake?
A tree is right about all eight things you checked. What does that promise about the next thing?
You can now take a machine apart with two questions. What was it shown, and who chose the rule?
Next comes Three Kinds of Guess, and Their Data.