CS344M Autonomous Multiagent Systems Patrick MacAlpine Department - - PowerPoint PPT Presentation

▶
cs344m autonomous multiagent systems
SMART_READER_LITE
LIVE PREVIEW

CS344M Autonomous Multiagent Systems Patrick MacAlpine Department - - PowerPoint PPT Presentation

Jan 05, 2024 350 likes •650 views

CS344M Autonomous Multiagent Systems Patrick MacAlpine Department or Computer Science The University of Texas at Austin Good Afternoon, Colleagues Are there any questions? Patrick MacAlpine Logistics How to read a research paper Patrick

slide-1
SLIDE 1

CS344M Autonomous Multiagent Systems

Patrick MacAlpine Department or Computer Science The University of Texas at Austin

slide-2
SLIDE 2

Good Afternoon, Colleagues

Are there any questions?

Patrick MacAlpine

slide-3
SLIDE 3

Logistics

  • How to read a research paper

Patrick MacAlpine

slide-4
SLIDE 4

Logistics

  • How to read a research paper

– Some have too few details...

Patrick MacAlpine

slide-5
SLIDE 5

Logistics

  • How to read a research paper

– Some have too few details... – Others have too many.

Patrick MacAlpine

slide-6
SLIDE 6

Logistics

  • How to read a research paper

– Some have too few details... – Others have too many.

  • Next week’s readings posted

Patrick MacAlpine

slide-7
SLIDE 7

Logistics

  • How to read a research paper

– Some have too few details... – Others have too many.

  • Next week’s readings posted

Patrick MacAlpine

slide-8
SLIDE 8

Overview of the Readings

Darwin: genetic programming approach

Patrick MacAlpine

slide-9
SLIDE 9

Overview of the Readings

Darwin: genetic programming approach Stone and McAllester: Architecture for action selection

Patrick MacAlpine

slide-10
SLIDE 10

Overview of the Readings

Darwin: genetic programming approach Stone and McAllester: Architecture for action selection Riley et al: Coach competition, extracting models

Patrick MacAlpine

slide-11
SLIDE 11

Overview of the Readings

Darwin: genetic programming approach Stone and McAllester: Architecture for action selection Riley et al: Coach competition, extracting models Kuhlmann et al: Learning for coaching

Patrick MacAlpine

slide-12
SLIDE 12

Overview of the Readings

Darwin: genetic programming approach Stone and McAllester: Architecture for action selection Riley et al: Coach competition, extracting models Kuhlmann et al: Learning for coaching Wihthop and Reidmiller: Reinforcement learning

Patrick MacAlpine

slide-13
SLIDE 13

Overview of the Readings

Darwin: genetic programming approach Stone and McAllester: Architecture for action selection Riley et al: Coach competition, extracting models Kuhlmann et al: Learning for coaching Wihthop and Reidmiller: Reinforcement learning MacAlpine, Price, and Stone: Role assignment

Patrick MacAlpine

slide-14
SLIDE 14

Overview of the Readings

Darwin: genetic programming approach Stone and McAllester: Architecture for action selection Riley et al: Coach competition, extracting models Kuhlmann et al: Learning for coaching Wihthop and Reidmiller: Reinforcement learning MacAlpine, Price, and Stone: Role assignment MacAlpine, Depinet, and Stone: Overlapping layered learning

Patrick MacAlpine

slide-15
SLIDE 15

Evolutionary Computation

  • Motivated by biological evolution: GA, GP

Patrick MacAlpine

slide-16
SLIDE 16

Evolutionary Computation

  • Motivated by biological evolution: GA, GP
  • Search through a space

Patrick MacAlpine

slide-17
SLIDE 17

Evolutionary Computation

  • Motivated by biological evolution: GA, GP
  • Search through a space

− Need a representation, fitness function − Probabilistically apply search operators to set of points in search space

Patrick MacAlpine

slide-18
SLIDE 18

Evolutionary Computation

  • Motivated by biological evolution: GA, GP
  • Search through a space

− Need a representation, fitness function − Probabilistically apply search operators to set of points in search space

  • Randomized, parallel hill-climbing through space

Patrick MacAlpine

slide-19
SLIDE 19

Evolutionary Computation

  • Motivated by biological evolution: GA, GP
  • Search through a space

− Need a representation, fitness function − Probabilistically apply search operators to set of points in search space

  • Randomized, parallel hill-climbing through space
  • Learning is an optimization problem (fitness)

Patrick MacAlpine

slide-20
SLIDE 20

Evolutionary Computation

  • Motivated by biological evolution: GA, GP
  • Search through a space

− Need a representation, fitness function − Probabilistically apply search operators to set of points in search space

  • Randomized, parallel hill-climbing through space
  • Learning is an optimization problem (fitness)

Some slides from Machine Learning [Mitchell, 1997]

Patrick MacAlpine

slide-21
SLIDE 21

Darwin United

  • More ambitious follow-up to Luke, 97 (made 2nd round)

Patrick MacAlpine

slide-22
SLIDE 22

Darwin United

  • More ambitious follow-up to Luke, 97 (made 2nd round)
  • Motivated in part by Peter’s detailed team construction

Patrick MacAlpine

slide-23
SLIDE 23

Darwin United

  • More ambitious follow-up to Luke, 97 (made 2nd round)
  • Motivated in part by Peter’s detailed team construction
  • Evolves whole teams — lexicographic fitness function

Patrick MacAlpine

slide-24
SLIDE 24

Darwin United

  • More ambitious follow-up to Luke, 97 (made 2nd round)
  • Motivated in part by Peter’s detailed team construction
  • Evolves whole teams — lexicographic fitness function
  • Lots of spinning, but figured out dribbling, offsides

Patrick MacAlpine

slide-25
SLIDE 25

Darwin United

  • More ambitious follow-up to Luke, 97 (made 2nd round)
  • Motivated in part by Peter’s detailed team construction
  • Evolves whole teams — lexicographic fitness function
  • Lots of spinning, but figured out dribbling, offsides
  • 1-1-1 record. Tied a good team, but didn’t advance

Patrick MacAlpine

slide-26
SLIDE 26

Darwin United

  • More ambitious follow-up to Luke, 97 (made 2nd round)
  • Motivated in part by Peter’s detailed team construction
  • Evolves whole teams — lexicographic fitness function
  • Lots of spinning, but figured out dribbling, offsides
  • 1-1-1 record. Tied a good team, but didn’t advance
  • Success of the method, but not pursued

Patrick MacAlpine

slide-27
SLIDE 27

Overlapping Layered Learning

  • Machine learning paradigms (not algorithms)

Patrick MacAlpine

slide-28
SLIDE 28

Overlapping Layered Learning

  • Machine learning paradigms (not algorithms)
  • Useful for learning complex skills that work well together

Patrick MacAlpine

slide-29
SLIDE 29

Overlapping Layered Learning

  • Machine learning paradigms (not algorithms)
  • Useful for learning complex skills that work well together
  • (slides)

Patrick MacAlpine