Information Elicitation in the Space-Time Module - PowerPoint PPT Presentation

About This Presentation
Title:

Information Elicitation in the Space-Time Module

Description:

Radar has only partial knowledge of the available resources and user needs. ... Useless info: There are no large rooms w/o a projector ... – PowerPoint PPT presentation

Number of Views:19
Avg rating:3.0/5.0
Slides: 10
Provided by: eugen8
Learn more at: http://www.cs.cmu.edu
Category:

less

Transcript and Presenter's Notes

Title: Information Elicitation in the Space-Time Module


1
Information Elicitationin the Space-Time Module
2
Information elicitation
  • Radar has only partial knowledge of the available
    resources and user needs.
  • It may have to elicit additional information from
    the users and administrator.
  • It needs to identify critical missing pieces of
    information and generate related questions.

3
Example Initial assignment
Available rooms
2
Roomnum. Area(feet2) Proj-ector
123 2,0001,0001,000 YesNoYes
1
3
  • Missing info
  • Invited talk Projector need
  • Poster session Room size Projector
    need
  • Assumptions
  • Invited talk Needs a projector
  • Poster session Small room is OK
    Needs no projector
  • Room requests
  • Invited talk, 910am Needs a large room
  • Poster session, 911am Needs a room

4
Example Choice of questions
Initial assignment
2
1
Posters
3
Talk
  • Candidate questions
  • Invited talk Needs a projector?
  • Poster session Needs a larger room? Needs
    a projector?
  • Room requests
  • Invited talk, 910am Needs a large room
  • Poster session, 911am Needs a room

5
Example Improved assignment
  • Room requests
  • Invited talk, 910am Needs a large room
  • Poster session, 911am Needs a room

Info elicitation
System Does the poster sessionneed a projector?
Posters
UserA projector may be useful,but not really
necessary.
6
Architecture
Top-level control
Processresource requests
. . .
7
Choice of questions
Step 1 Fast heuristic pruning ofunimportant
questions.
Step 2 Slower selection of the most relevant
questions.
8
Choice of questions
  • Step 1 Fast heuristic pruning
  • For each candidate question, estimate the
    probabilities of possible answers.
  • For each possible answer, compute the respective
    change of the assignment utility.
  • For each question, compute its expected impact on
    the utility, and prune questions with small
    expected impacts.

9
Choice of questions
  • Step 2 Selection of relevant questions
  • For each remaining question and each possible
    answer, call the optimizer to search for a better
    assignment, and determine the resulting utility
    increase.
  • Prune the questions that do not lead to
    significant utility increase.
  • Send the remaining questions to the users.
Write a Comment
User Comments (0)
About PowerShow.com