Bayesian Decision Analysis Quiz 4

JavaScript isn't enabled in your browser, so this file can't be opened. Enable and reload.

How does the `update` function compute the likelihood of the data?

1 point

It adds up the unnormalized posteriors.

It uses the binomial distribution.

It uses the prior probabilities in the pmf.

It uses the index of the pmf, which contains hypothetical probabilities of winning.

Clear selection

Which of the following statements about the `choose` function are true?

2 points

It uses `pmf_choice` to choose a random value from each posterior distribution.

It uses `argmax` to find the index of the distribution that yielded the highest value.

The chance of choosing each machine is proportional to its posterior mean.

Before we see any data, it chooses each machine with the same probability.

It implements a strategy called Thompson sampling.

If we run the Bayesian bandit strategy for a while, which of the following are likely to be true?

2 points

All machines get played equally often.

The best machines get played more often than the others.

The posterior mean for the best machines is higher than for the others.

The credible interval for the best machines is wider than for the others.

The credible interval for most machines contains the actual probability of winning.

Submit

Clear form

This content is neither created nor endorsed by Google. Report Abuse - Terms of Service - Privacy Policy