# Lecture 13 Examples

📗 My handwriting is really bad, you should copy down your notes from the lecture videos instead of using these.
Lecture 13 Zoom Annotated (2021): Link
Lecture 13 Pre-recorded Annotated (from 2020, please use with caution): Link

# Q13 Quiz Instruction

📗 The quiz is canceled. Everyone gets 0.5 points.

# Multi-Armed Bandit Demo

This is a demo for multi-armed bandit. The boxes have different mean rewards between 0 and 1. Click on one of them to collect the reward. For TopHat quiz questions, enter code and press Reset.



Pull All Pull Random Pull Highest Mean Pull Highest UCB
Number of arm pulls per round:
Remaining number of pulls:
📗 Bandit Settings: Reset
(1) Number of arms:
(2) Mean reward: Random Uniform Close
(3) Reward distribution: Gaussian Uniform , Standard deviation (or half range):
📗 Algorithm Settings: Run Algorithm
(1) Number of arm pulls:
(2) Algorithm: Greedy EXP UCB , \(\varepsilon\) (for Greedy and EXP: , c (for UCB):
📗 Output:
(1) Data:
(2) Means:
(3) Counts:
(4) Total reward:
(5) Total regret: