chrispiech / probabilityforcomputerscientists Goto Github PK

Page: https://chrispiech.github.io/probabilityForComputerScientists/en/part2/exponential/

The earthquake example uses a "magnitude 8.0+". This seems unusually high. This doesn't affect the example in anyway. But, can we use a more reasonable magnitude of say 7.0+?

Many paragraphs are missing p tags. This becomes an issue on mobile!

Its a bit too contentious...

It would be so cool. Use material from this quarters inference 2 lecture

The following are all cited by students to mean the same thing. Perhaps they could be consolidated into one term.

• Product Rule of Counting
• Step Rule of Counting
• General Principle of Counting
• Fundamental Principle of Counting
• Distinct buckets method of Counting

It would be great to show a proof for this claim:
This is equivalent mathematically to calculating the posterior for one observation and calling the posterior your new prior

in the Digital Eye Test example

Page: https://chrispiech.github.io/probabilityForComputerScientists/en/part2/normal/

States the following:

By definition a Normal has $E[X] = \mu$.

In fact this equality requires a proof. The proof can be seen here.

https://statproofbook.github.io/P/norm-mean.html

I do not see how this equality can be asserted from any kind of definition without making huge leaps of assumptions.

In Part 2: Random Variables, we should have a section on drawing a sample from a random variable! It is such a useful concept (which we refer to many times). We could include inverse box sampling as a neat algorithm

When cancel the selecting formula, there's a blank over the formula.

This demo needs a new figure and a lot more teaching:
en/examples/dart_logo/

https://chrispiech.github.io/probabilityForComputerScientists/en/part1/probability/

en/part3/multinomial/

explain the origin of the PMF similar to multiple coin flips example

Page: https://chrispiech.github.io/probabilityForComputerScientists/en/part2/exponential/

Poisson is stated in lower case. For example: "It assumes that the events occur via a poisson process." Is this a common thing or an error?

Importantly it doesn't include the definition of P(EF) = P(E)P(F)

The categorical is a multinomial with a single "dice" roll
https://en.wikipedia.org/wiki/Categorical_distribution

It often comes up in student questions, implicitly. It might be work including as a random variable!