# Probabilistic Foundations of Artificial Intelligence

## Overview

How can we build systems that perform well in uncertain environments and unforeseen situations? How can we develop systems that exhibit ''intelligent'' behavior, without prescribing explicit rules? How can we build systems that learn from experience in order to improve their performance? We will study core modeling techniques and algorithms from statistics, optimization, planning, and control and study applications in areas such as sensor networks, robotics, and the Internet. The course is designed for upper-level undergraduate and graduate students.## Topics covered

- Tutorial in logic (propositional, first-order)
- Probability
- Bayesian Networks (models, exact and approximate inference, learning)
- Temporal models (Hidden Markov Models, Dynamic Bayesian Networks)
- Probabilistic planning (MDPs, POMDPs)
- Reinforcement learning
- Combining logic and probability

## News

- Exam review session on Friday, Jan 16, 13:00-15.00 in CHN C 14.
- No classes and recitations on Friday, Dec 12.
- More office hours on Thursdays, Nov 13 and Nov 20, 14.00-16.00 in CAB E 62.1.
- Office hours for Bayesian network programming assignments on Thursdays Oct 30 and Nov 6, 14.00-16.00 in CAB E 62.1.

## Details

**VVZ Information:**See here.**Lecture:**Friday 10-12 in CHN C 14**Recitations:**Friday 13-14 (Last names A-L) and 14-15 (Last names M-Z) in CHN C 14**Teaching assistants:**- Yuxin Chen [yuxin (dot) chen (at) inf (dot) ethz (dot) ch]
- Carlos Cotrini [ccarlos (at) inf (dot) ethz (dot) ch]
- Josip Djolonga [josipd (at) inf (dot) ethz (dot) ch]
- Alkis Gotovos [alkisg (at) inf (dot) ethz (dot) ch]
- Baharan Mirzasoleiman [baharanm (at) inf (dot) ethz (dot) ch]
- Hastagiri Vanchinathan [hastagiri (at) inf (dot) ethz (dot) ch]

**Textbook:**S. Russell, P. Norvig. Artificial Intelligence: A Modern Approach (3rd Edition).

## Homeworks

- Homework 1 [pdf] — Solutions 1 [pdf]
- Homework 2 [pdf] — Solutions 2 [pdf]
- Homework 3 [pdf] [zip] — Solutions 3 [zip]
- Homework 4 [pdf] [zip] — Solutions 4 [pdf] [zip]
- Homework 5 [pdf] [zip] — Solutions 5 [pdf] [zip]
- Homework 6 [pdf] — Solutions 6 [pdf]
- Homework 7 [pdf] [zip] — Solutions 7 [pdf]

## Lecture Notes

- Sep 19: Introduction, Propositional logic (Chapters 1 and 7.1-8.3)
- Sep 26: Propositional Resolution, First-order Logic (Chapters 7.1-8.3, Chapters 9.1-9.2, 9.5)
- Oct 3: Inference in First Order Logic; Intro to probability (Chapters 9.1-9.2, 9.5, 13.1-5)
- Oct 10: Bayesian Networks (Chapters 14.1-14.2); JavaBayes applet
- Oct 17: Bayesian Networks: D-separation (Chapters 14.1-14.2); JavaBayes applet
- Oct 24: Inference in Bayesian Networks (Chapters 14.4)
- Oct 31: Approximate Inference (Chapter 14.5; Bishop Chapter 8.4)
- Nov 7: Sampling, MCMC (Chapter 14.5; Bishop Chapter 8.4)
- Nov 14: Temporal models: HMMs, Kalman Filters (Chapter 15.1-15.5)
- Nov 21: Particle filtering. Probabilistic Planning: MDPs (Chapter 17.1-17.3).
- Nov 28: Probabilistic Planning: MDPs (Chapter 20.1-2). Value iteration demo (at UBC)
- Dec 5: Learning Bayesian Networks (Chapter 20.1-2)
- Dec 19: Reinforcement learning (Chapter 21.1-3). Q-learning demo (at Northwestern).

## Recitations

- Sep 26: Logic [pdf]
- Oct 3: Propositional logic [pdf], HW1 out
- Oct 10: Probability review, Intro to Bayesian networks, HW2 out
- Oct 17: d-separation, HW1 solutions, HW3 out
- Oct 24: HW2 solutions, HW4 out (more details next week)
- Oct 31: Belief propagation, more details on HW3 and HW4
- Nov 7: Gibbs sampling
- Nov 14: More on MCMC, HW5 out
- Nov 21: Temporal models [pdf]
- Nov 28: MDPs, HW6 out
- Dec 5: Learning Bayes nets, HW6 solutions, HW7 out
- Dec 19: Reinforcement learning [pdf], HW7 solutions

## Old Exams

## Relevant Readings

- Christopher M. Bishop. Pattern Recognition and Machine Learning. Springer, 2007 (optional)