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

Info iconThis preview shows pages 1–6. Sign up to view the full content.

View Full Document Right Arrow Icon
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 s Course contest s Fun! (And extra credit.) s Regular tournaments s Instructions posted soon! 2
Background image of page 1

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

View Full DocumentRight Arrow Icon
2 Mid-Semester Evals s Generally, things seem good! s General s Examples are appreciated in lecture s Favorite aspect: projects (almost all) --- writtens significantly less preferred s Office hours: s Most common answers: “Helpful.” and “Haven’t gone.” s Some: too crowded. b perhaps try a different office hour slot s Section: s Split between basically positive and don’t go s Assignments s Written: median time 6hrs s Programming: median time 10hrs s Exams: s Midterm: evening (13) vs in-class (11) or indifferent (8) s Do the contest 3 Outline s HMMs: representation s HMMs: inference s Forward algorithm s Particle filtering 8
Background image of page 2
3 Recap: Reasoning Over Time s Stationary Markov models s 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 s HMMs have two important independence properties: s Markov hidden process, future depends on past via the present s Current observation independent of all else given current state s Quiz: does this mean that observations are independent given no evidence? s [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
Background image of page 3

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

View Full DocumentRight Arrow Icon
4 Real HMM Examples s Speech recognition HMMs: s Observations are acoustic signals (continuous valued) s States are specific positions in specific words (so, tens of thousands) s Machine translation HMMs: s Observations are words (tens of thousands) s States are translation options s Robot tracking: s Observations are range readings (continuous) s States are positions on a map (continuous) Outline s HMMs: representation s HMMs: inference s Forward algorithm s Particle filtering 12
Background image of page 4
5 Filtering / Monitoring s
Background image of page 5

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

View Full DocumentRight Arrow Icon
Image of page 6
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 04/21/2010 for the course EECS 188 taught by Professor Cs188 during the Spring '01 term at University of California, Berkeley.

Page1 / 18

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

This preview shows document pages 1 - 6. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online