{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

lecture 20

# lecture 20 - CS 188 Artificial Intelligence Spring 2010...

This preview shows pages 1–5. Sign up to view the full content.

1 CS 188: Artificial Intelligence Spring 2010 Lecture 20: HMMs and Particle Filtering 4/5/2010 Pieter Abbeel --- UC Berkeley Many slides over this course adapted from Dan Klein, Stuart Russell, Andrew Moore Announcements square4 Course contest square4 Fun! (And extra credit.) square4 Regular tournaments square4 Instructions posted soon! 2

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
2 Mid-Semester Evals square4 Generally, things seem good! square4 General square4 Examples are appreciated in lecture square4 Favorite aspect: projects (almost all) --- writtens significantly less preferred square4 Office hours: square4 Most common answers: “Helpful.” and “Haven’t gone.” square4 Some: too crowded. barb2right perhaps try a different office hour slot square4 Section: square4 Split between basically positive and don’t go square4 Assignments square4 Written: median time 6hrs square4 Programming: median time 10hrs square4 Exams: square4 Midterm: evening (13) vs in-class (11) or indifferent (8) square4 Do the contest 3 Outline square4 HMMs: representation square4 HMMs: inference square4 Forward algorithm square4 Particle filtering 8
3 Recap: Reasoning Over Time square4 Stationary Markov models square4 Hidden Markov models X 2 X 1 X 3 X 4 rain sun 0.7 0.7 0.3 0.3 X 5 X 2 E 1 X 1 X 3 X 4 E 2 E 3 E 4 E 5 X E P rain umbrella 0.9 rain no umbrella 0.1 sun umbrella 0.2 sun no umbrella 0.8 Conditional Independence square4 HMMs have two important independence properties: square4 Markov hidden process, future depends on past via the present square4 Current observation independent of all else given current state square4 Quiz: does this mean that observations are independent given no evidence? square4 [No, correlated by the hidden state] X 5 X 2 E 1 X 1 X 3 X 4 E 2 E 3 E 4 E 5

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
4 Real HMM Examples square4 Speech recognition HMMs: square4 Observations are acoustic signals (continuous valued) square4 States are specific positions in specific words (so, tens of thousands) square4 Machine translation HMMs: square4 Observations are words (tens of thousands) square4 States are translation options square4 Robot tracking: square4 Observations are range readings (continuous) square4 States are positions on a map (continuous) Outline square4 HMMs: representation square4 HMMs: inference square4 Forward algorithm square4 Particle filtering 12
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

### What students are saying

• As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

Kiran Temple University Fox School of Business ‘17, Course Hero Intern

• I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

Dana University of Pennsylvania ‘17, Course Hero Intern

• The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

Jill Tulane University ‘16, Course Hero Intern