View Full Document

BLOG: Probabilistic Models with Unknown Objects



View the full content.
View Full Document
View Full Document

4 views

Unformatted text preview:

1 BLOG Probabilistic Models with Unknown Objects Brian Milch Computer Science Division University of California at Berkeley USA milch cs berkeley edu http www cs berkeley edu milch Bhaskara Marthi Computer Science Division University of California at Berkeley USA bhaskara cs berkeley edu http www cs berkeley edu bhaskara Stuart Russell Computer Science Division University of California at Berkeley USA russell cs berkeley edu http www cs berkeley edu russell David Sontag Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology USA dstontag csail mit edu http people csail mit edu dsontag Daniel L Ong Computer Science Division University of California at Berkeley USA dlong ocf berkeley edu http www ocf berkeley edu dlong Andrey Kolobov Computer Science Division University of California at Berkeley USA karaya1 rambler ru 2 1 1 BLOG Probabilistic Models with Unknown Objects Introduction Human beings and AI systems must convert sensory input into some understanding of what is going on in the world around them That is they must make inferences about the objects and events that underlie their observations No pre specified list of objects is given the agent must infer the existence of objects that were not known initially to exist In many AI systems this problem of unknown objects is engineered away or resolved in a preprocessing step However there are important applications where the problem is unavoidable Population estimation for example involves counting a population by sampling from it randomly and measuring how often the same object is resampled this would be pointless if the set of objects were known in advance Record linkage a task undertaken by an industry of more than 300 companies involves matching entries across multiple databases These companies exist because of uncertainty about the mapping from observations to underlying objects Finally multi target tracking systems perform data association connecting say radar blips to



Access the best Study Guides, Lecture Notes and Practice Exams

Loading Unlocking...
Login

Join to view BLOG: Probabilistic Models with Unknown Objects and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view BLOG: Probabilistic Models with Unknown Objects and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?