CS 188 Section Materials, Fall 2016
Introduction to Artificial Intelligence
Discussions: Tues 2-3 in 310 Soda, Wed 1-2 in Etcheverry 3113
Office Hours: Fri 1-3 in 341A Soda
Join the section mailing list by clicking here
Introduction to Artificial Intelligence
Discussions: Tues 2-3 in 310 Soda, Wed 1-2 in Etcheverry 3113
Office Hours: Fri 1-3 in 341A Soda
Join the section mailing list by clicking here
Extra Practice Materials
I took this class a student in Fall 2015 with Pat Virtue and Stuart Russell. Here are some of my old assignments.
You are welcome to practice with these. Our syllabus was a bit different back then, so it's up to you to find the relevant problems.
Section Slideshows/Notes
Supplementary Materials
Random Useful (?) Things
Here is the Spotify playlist I use during Berkeley time.
I highly recommend getting familiar with pdb this semester (the Python debugger). It's very simple to use: all you need to know is pdb.set_trace() to get started. Debuggers are way better than using print statements to debug, and this will probably save you a lot of headache in the projects this semester. Here's the doc page for pdb. If you ever need help with pdb, feel free to email me or ask in Office Hours! :)
You will find Git useful this semester. If you need a refresher or have never used Git, here's a great guide from last semester's 61B.
If you're looking for a new code editor (or you're tired of Sublime Text telling you to buy it), try Visual Studio Code. It has most of the nice features and looks of Sublime, and it comes with really awesome Git features that will be useful in group projects (like in 188).
This is a really nice article written by Stuart Russell on why we shouldn't give robots guns.
I took this class a student in Fall 2015 with Pat Virtue and Stuart Russell. Here are some of my old assignments.
You are welcome to practice with these. Our syllabus was a bit different back then, so it's up to you to find the relevant problems.
- Fall 2015 Final
- Fall 2015 HW6: Bayes Nets and Variable Elimination
- Fall 2015 HW8: HMM, DBN, and Decision Networks
- Fall 2015 Section 7: HMM
Section Slideshows/Notes
- Discussion 0: State Space and Search
- Slideshow
- Lecture notes I used for search (Mainly for me. Hopefully helps you.)
- Discussion 1: More Search, plus Heuristics
- Discussion 2: CSPs and Game Trees
- Discussion 3: Game Trees, Utility, MDPs
- Discussion 4: MDPs, Q-Learning
- Slideshow
- Note: I wasn't able to teach this week. The slideshow has some high level things on Q-Learning to get you primed for understanding the lectures. See you next week!
- Discussion 5: Reinforcement Learning
- Discussion 6: Probability and Bayes Nets
- Discussion 7: Bayes Net Sampling, Variable Elimination
- Slideshow (Now a Google Doc that you can comment on!)
- Discussion 8: VPI
- Discussion 9: HMM
- Discussion 10: Machine Learning: Naive Bayes and Perceptron
- Discussion 11: Neural Nets and Computation Graphs
- Did a quick video on Discussion 11. This was all in one take, so it's very rough. Please read the erata if you chose to watch.
- Worksheet and Solutions if you want to follow along.
- Overview: https://www.youtube.com/watch?v=fOztvXo3H00
- Q1: https://www.youtube.com/watch?v=UCuZ5aqigOU
- Erata: At the 2:20 mark, I say "Pre-activation Function" instead of "Pre-activation value."
- Q3: https://www.youtube.com/watch?v=n3dSGX8apXc
- Erata: At the 3 min mark, I say Neural Net, when I should have said neuron.
- At the 4 min mark, I verbally flip the subtraction (though we square so it's OK by complete luck...)
- Q4: https://www.youtube.com/watch?v=ETj6qQcl0pc
- Erata: 5:27 mark: We only take the derivate with "respect to" one variable, not both as implied by my slip.
- Also: thank you for a great semester! It was a joy to TA for y'all.
Supplementary Materials
- Download for the Backtracking Visual shown in Lecture.
- This is one of the best animations I've seen for understanding Alpha Beta Pruning.
- Probability (A comprehensive set of notes on probability in the style we deal with it in this class)
Random Useful (?) Things
Here is the Spotify playlist I use during Berkeley time.
I highly recommend getting familiar with pdb this semester (the Python debugger). It's very simple to use: all you need to know is pdb.set_trace() to get started. Debuggers are way better than using print statements to debug, and this will probably save you a lot of headache in the projects this semester. Here's the doc page for pdb. If you ever need help with pdb, feel free to email me or ask in Office Hours! :)
You will find Git useful this semester. If you need a refresher or have never used Git, here's a great guide from last semester's 61B.
If you're looking for a new code editor (or you're tired of Sublime Text telling you to buy it), try Visual Studio Code. It has most of the nice features and looks of Sublime, and it comes with really awesome Git features that will be useful in group projects (like in 188).
This is a really nice article written by Stuart Russell on why we shouldn't give robots guns.