More properhes of vpi acquiring informahon ophmally is

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: given new informaHon. 10 How InformaHon is Doled Out •  VPI = Value of Perfect InformaHon •  In pracHce, informaHon is: •  ParHal •  Imperfect •  ParHal informaHon: •  We learn about some state variables, but don’t learn the exact state of the world •  Example: We can see a traffic camera at one intersecHon, but we don’t have coverage of our enHre route •  Imperfect informaHon: •  We learning something that may not be reliable •  Example: There may be a lag in our traffic data •  Our framework can handle this by introducing an extra variable. (We get perfect informaHon about the observed variable, and this influences the distribuHon over the others.) ȹ ȹ VPIE (E ) = ȹ ∑ P(E ' | E )EUE ,E + (a' | E, E ' )ȹ − EU(a | E ) + ȹ E + Ⱥ € Examples Where Value of InformaHon is (should be) Considered •  •  •  •  •  •  Medical tests (x ­rays, CT ­scans, mammograms, etc.) Pregnancy tests Pre ­purchase house/car inspecHons Subscribing to Consumer Reports Hiring consultants Hiring a trainer •  •  •  •  •  •  •  Funding research Checking one’s own credit scor...
View Full Document

Ask a homework question - tutors are online