RadarSpaceTime

1
RADAR
Automated Assistantfor Crisis Management
(Reflective Agent withDistributed Adaptive
Reasoning)
www.radar.cs.cmu.edu
2
Purpose

• Automation of office-management tasks, such as
  scheduling, e-mail handling, and resource
  allocation

3
Outline

• Overview of RADAR
• Resource allocation
• Future challenges

• More information
• See www.radar.cs.cmu.edu
• Talk with Radar researchers

4
Outline

• Overview of RADAR
• Resource allocation
• Future challenges

5
PAL video
Three-minute videoMilitary-setting motivation
forRADAR (Carnegie Mellon)and CALO (SRI).
6
Project size

• Largest research project in CMUs
• School of Computer Science.

• Five departmentsLanguage Technologies
  (LTI)Computer Science Department (CSD)Institute
  for Software Research International
  (ISRI)Human-Computer Interaction Institute
  (HCII)Center for Automated Learning and
  Discovery (CALD)

• Eighty peopleTwenty-nine faculty
  membersTwenty-seven graduate studentsTwenty-four
  others

• Five years (20032008)

7
Project size

• Largest research project in CMUs
• School of Computer Science.

• Advantages
• Multiple research areas
• Collaboration opportunities
• Potential of a major impact

• Drawbacks
• Coordination challenges
• Frequent deliverables

8
Challenges

• Intelligent performance ofoffice-management tasks

• Collaboration with ahuman administrator

• Dialog with users

• Continuous learning of new knowledge and
  strategies

• Integration of multiple tasks

9
Research areas

• Artificial intelligence

• Machine learning

• Natural-language processing

• Human-computer interaction

• Architecture and integration

10
Main components
Planning and co-ordination of the systems
high-level actions.
11
Main components
Web Master
Helps create and maintain web sites.
12
Main components
Web Master
E-MailOrganizer
Helps filter, sort, and prioritize messages.
13
Main components
Web Master
E-MailOrganizer
Calendar Manager
Helps keep track of appointments and negotiate
meeting times among multiple users.
14
Main components
Web Master
E-MailOrganizer
Calendar Manager
Briefing Assistant
Helps compile reports based on multiple data
sources.
15
Main components
Resource Allocation
16
Outline

• Overview of RADAR
• Resource allocation
• Future challenges

17
Purpose

• Automated allocation of office
• resources, in both routine and
• crisis situations.

18
People
Faculty
Jaime Carbonelljgc_at_cs.cmu.eduResource
allocation(AI and learning)
Eugene Finke.fink_at_cs.cmu.eduResource
allocation(AI and learning)
Bob Frederkingref_at_cs.cmu.eduE-mail
understanding(Natural language)
19
RADAR/Space video
Six-minute video Initial system for
automated assignment of offices.
20
Initial results

• A prototype system for automated
• allocation of offices.

• Effective allocation of office resources
• Processing of natural-language requests
• Interface for a human administrator

21
Outline

• Overview of RADAR
• Resource allocation
• Future challenges

22
Motivating task

• Scheduling of talks at a conference,
• and related allocation of rooms and
• equipment, in a crisis situation.

23
Automated reasoning

• Temporal reasoning

• Uncertainty tolerance

• Preference elicitation

• Collaboration with ahuman administrator

24
Learning

• Integrated learning of new
• knowledge and strategies.

• From experience
• From observation
• From instruction

25
Integration
Task manager
High-level planningIntegrated learning
RADAR CalendarManager
RADAR E-MailOrganizer
RADAR WebMaster
RADAR ResourceAllocation
RADAR BriefingAssistant
Users
Integrated RADAR
26
Integration
Knowledgebase andinferences
High-level planningIntegrated learning
RADAR ResourceAllocation
Users
Integrated RADAR
27
Tasks and skills

• Development of AI, learning, and natural-language
  algorithms

• Solving open-ended problems

• Implementation and integration

28
More information
AI and learning
Jaime Carbonellwww.cs.cmu.edu/jgcjgc_at_cs.cmu.edu
Newell Simon Hall 4519
www.radar.cs.cmu.edu