Probability Theory lecture, various TODOs

[fonsp]

From #61 by @bertdv

I made some minor updates to PT. Here is what I still like to be done:

• in cell Code Example: Sum of Two Gaussian-distributed Variables, can we make the parameters change-able w sliders?

• In cell Revisiting the Challenge: Disease Diagnosis, can we make the assumptions (true positive rate etc) adaptable?

• In Code Example: Sampling Distribution and Likelihood Function for the Coin Toss, can we make the assumptions adaptable?

• make a "pluto box" for a CHALLENGE (with its own color), and apply to the Challenge: Disease diagnosis cell.

  • Also apply to the cell Revisiting the Challenge: Disease Diagnosis
  • Apply the Challenge box to Inference Exercise: Bag Counter cell

• make a "pluto box" for IMPORTANT , an apply to to the rightarrow Probability theory (PT) provides the **theory of optimal processing of incomplete information** statement

  • Add IMPORTANT box to cell All legitimate probabilistic relations can be derived from the sum and product rules!
  • Add IMPORTANT box to cell Bayes rule is the fundamental rule for learning from data!

• HTML does not seem to render in pluto-markdown, see <span style="font-size:large; color:red"> Bayes rule is the fundamental rule for learning from data! </span> cell (what is wrong)?

• Make a CODE EXAMPLE pluto box, and apply to Code Example: Sampling Distribution and Likelihood Function for the Coin Toss cell

  • Also apply to cell Code Example: Sum of Two Gaussian Distributed Variables

• (Bert to do) end with a summary of key equations.

[fonsp]

Fixed in #154