fa13-cs188-lecture-17-1PP

Over evidence sets probabilisc reasoning needed to

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: eport POMDPs POMDPs s   MDPs have:         States S Ac)ons A Transi)on func)on P(s |s,a) (or T(s,a,s )) Rewards R(s,a,s ) a s, a s,a,s s   POMDPs add: b   Observa)ons O   Observa)on func)on P(o|s) (or O(s,o))   POMDPs are MDPs over belief states b (distribu)ons over S)   We ll be able to say more in a few lectures a b, a o b Example: Ghostbusters   In (sta)c) Ghostbusters:   Belief state determined by evidence to date {e}   Tree really over evidence sets   Probabilis)c reasoning needed to predict new evidence given past evidence   Solving POMDPs   One way: use truncated expec)max to compute approximate value of ac)ons   What if you only considered bus)ng or one sense followed by a bust?   You get a VPI ­ba...
View Full Document

This note was uploaded on 12/22/2013 for the course CS 188 taught by Professor Staff during the Fall '08 term at University of California, Berkeley.

Ask a homework question - tutors are online