CS344M Autonomous Multiagent Systems Todd Hester Department of - - PowerPoint PPT Presentation

cs344m autonomous multiagent systems
SMART_READER_LITE
LIVE PREVIEW

CS344M Autonomous Multiagent Systems Todd Hester Department of - - PowerPoint PPT Presentation

CS344M Autonomous Multiagent Systems Todd Hester Department of Computer Science The University of Texas at Austin Good Afternoon, Colleagues Are there any questions? Todd Hester Logistics All readings up Todd Hester Logistics All


slide-1
SLIDE 1

CS344M Autonomous Multiagent Systems

Todd Hester Department of Computer Science The University of Texas at Austin

slide-2
SLIDE 2

Good Afternoon, Colleagues

Are there any questions?

Todd Hester

slide-3
SLIDE 3

Logistics

  • All readings up

Todd Hester

slide-4
SLIDE 4

Logistics

  • All readings up
  • More reflections on peer reviews?

Todd Hester

slide-5
SLIDE 5

Logistics

  • All readings up
  • More reflections on peer reviews?
  • Final projects due in 2 weeks!

Todd Hester

slide-6
SLIDE 6

Reading Overview: Vidal and Durfee

Recursive Modeling Method

  • What should I do?

Todd Hester

slide-7
SLIDE 7

Reading Overview: Vidal and Durfee

Recursive Modeling Method

  • What should I do?
  • What should I do given what I think you’ll do?

Todd Hester

slide-8
SLIDE 8

Reading Overview: Vidal and Durfee

Recursive Modeling Method

  • What should I do?
  • What should I do given what I think you’ll do?
  • What should I think you’ll do given what I think you think I’ll

do?

Todd Hester

slide-9
SLIDE 9

Reading Overview: Vidal and Durfee

Recursive Modeling Method

  • What should I do?
  • What should I do given what I think you’ll do?
  • What should I think you’ll do given what I think you think I’ll

do?

  • etc.

Todd Hester

slide-10
SLIDE 10

Prediction Method

  • Situations

− Includes physical and mental states

Todd Hester

slide-11
SLIDE 11

Prediction Method

  • Situations

− Includes physical and mental states − Could be computationally expensive

  • Trade-off between time and performance gain

Todd Hester

slide-12
SLIDE 12

Prediction Method

  • Situations

− Includes physical and mental states − Could be computationally expensive

  • Trade-off between time and performance gain
  • When is it worthwhile to model deeper?

Todd Hester

slide-13
SLIDE 13

Lessons

  • Modeling can help
  • There is a lot of useless information in recursive models
  • Approximations (limited rationality) can be useful

Todd Hester

slide-14
SLIDE 14

Learning Teammate Models for Ad Hoc Teamwork

  • Forced to work with a group of unknown teammates

Todd Hester

slide-15
SLIDE 15

Learning Teammate Models for Ad Hoc Teamwork

  • Forced to work with a group of unknown teammates
  • Start with learned models of prior teammates

Todd Hester

slide-16
SLIDE 16

Learning Teammate Models for Ad Hoc Teamwork

  • Forced to work with a group of unknown teammates
  • Start with learned models of prior teammates
  • Plan using these models to perform well on task

Todd Hester

slide-17
SLIDE 17

Learning Teammate Models for Ad Hoc Teamwork

  • Forced to work with a group of unknown teammates
  • Start with learned models of prior teammates
  • Plan using these models to perform well on task
  • Slides...

Todd Hester

slide-18
SLIDE 18

Where do Models Come From?

Observation:

  • RMM: Use existing model
  • Barrett: Learn model from prior experiences

What if we can’t build a full model in advance?

Todd Hester

slide-19
SLIDE 19

Where do Models Come From?

Observation:

  • RMM: Use existing model
  • Barrett: Learn model from prior experiences

What if we can’t build a full model in advance?

  • How can we build a predictive model on-line incrementally?

Todd Hester

slide-20
SLIDE 20

Play me at RoShamBo

  • Rock beats scissors
  • Scissors beats paper
  • Paper beats rock

Todd Hester

slide-21
SLIDE 21

Play me at RoShamBo

  • Rock beats scissors
  • Scissors beats paper
  • Paper beats rock
  • What is your strategy before modeling me?

Todd Hester

slide-22
SLIDE 22

Play me at RoShamBo

  • Rock beats scissors
  • Scissors beats paper
  • Paper beats rock
  • What is your strategy before modeling me?
  • What is your strategy after modeling me?

Todd Hester

slide-23
SLIDE 23

Play me at RoShamBo

  • Rock beats scissors
  • Scissors beats paper
  • Paper beats rock
  • What is your strategy before modeling me?
  • What is your strategy after modeling me?
  • Am I modeling you?

Todd Hester

slide-24
SLIDE 24

Play me at RoShamBo

  • Rock beats scissors
  • Scissors beats paper
  • Paper beats rock
  • What is your strategy before modeling me?
  • What is your strategy after modeling me?
  • Am I modeling you?

Todd Hester

slide-25
SLIDE 25

Discussion

  • How do you deal with a teammate/opponent who is

adapting to you as well?

Todd Hester

slide-26
SLIDE 26

Discussion

  • How do you deal with a teammate/opponent who is

adapting to you as well?

  • Applications of ad hoc teamwork?

Todd Hester

slide-27
SLIDE 27

Discussion

  • How do you deal with a teammate/opponent who is

adapting to you as well?

  • Applications of ad hoc teamwork?
  • What if there was communication?

Todd Hester

slide-28
SLIDE 28

Discussion

  • How do you deal with a teammate/opponent who is

adapting to you as well?

  • Applications of ad hoc teamwork?
  • What if there was communication?
  • How would you build an ad hoc teammate?

Todd Hester