SlideShare una empresa de Scribd logo
1 de 21
Descargar para leer sin conexión
Robot Swarms as Ensembles of
Cooperating Components
Matthias Hölzl
With contributions from Martin Wirsing, Annabelle Klarl
AWASS
Lucca, June 24, 2013
www.ascens-ist.eu
The Task
Robots cleaning an exhibition area
Matthias Hölzl 2
marXbot
Miniature mobile robot developed by EPFL
Rough-terrain mobility
Robots can dock to other robots
Many sensors
Proximity sensors
Gyroscope
3D accelerometer
RFID reader
Cameras
. . .
ARM-based Linux system
Gripper for picking up items
Matthias Hölzl 3
Swarm Robotics
Matthias Hölzl 4
Problems
Noise, sensor resolution
Extracting information from sensor data
Unforeseen situations
Uncertainty about the environment
Performing complex actions when intermediate
results are uncertain
. . .
Matthias Hölzl 5
Action Logics
Logics that can represent change over time
Probabilistic behavior can be modeled (but is cumbersome)
Matthias Hölzl 6
Markov Decision Processes
pos = (x,y)
pos = (x,y+1) pos = (x+1,y+1)
pos = (x + 1,y)
e / ...
s, n / ...
s / 0.9 / -0.1
e,w / 0.025 / -0.1
w / ...
s,n / ...
e / ...
s, n / ...
w/ ...
s,n / ...
n / 0.9 / -0.1
e, w / 0.025 / -0.1
s,n,w,e / 0.05 / -0.1
Matthias Hölzl 7
Markov Decision Processes
watchTV
goToClub
Decide
Activity
In Club
Watching
TV
Dancing
Alone
Drinking
Dancing
With
Partner
dance p = 0.5
dance p = 0.5
drinkBeer
Oh, oh
Oh, no
flirt
p = 0.05
p = 0.95
Matthias Hölzl 8
MDPs: Strategies
watchTV
goToClub
Decide
Activity
In Club
Watching
TV
Dancing
Alone
Drinking
Dancing
With
Partner
dance p = 0.5
dance p = 0.5
drinkBeer
Oh, oh
Oh, no
flirt
p = 0.05
p = 0.95
State TV CDB CDF
DA watchTV goToClub goToClub
IC drinkBeer drinkBeer dance
DWP flirt flirt flirt
Utility 0.1 −0.05(∗) −1.975(∗)
(∗) 0.05 + (−0.1)
(∗∗) 0.05 + (0.5 × 0.2) + 0.5 × (0.25 + (0.05 × 5) + (0.95 × −5))
Matthias Hölzl 9
Reinforcement Learning
General idea:
Figure out the expected
value of each action
in each state
Pick the action with
the highest expected
value (most of the
time)
Update the expectations
according to the actual
rewards
Matthias Hölzl 10
How well does this work?
Rather well for small problems
But: state explosion
Matthias Hölzl 11
Solutions
Decomposition
Hierarchy
Partial programs
Matthias Hölzl 12
POEM
Action language
First-order reasoning
Hierarchical reinforcement learning
Learns completions for partial programs
Concurrency
Reflection / meta-object protocols
· · ·
Matthias Hölzl 13
Iliad: A POEM Implementation
Common Lisp-based programming language
Full first-order reasoning
Operations on logical theories: U(C)NA, domain closure, . . .
Resolution, hyperresolution, DPLL, etc.
Conditional answer extraction
Procedural attachment, constraint solving
Hierarchical reinforcement learning
Based on Concurrent ALisp
Partial programs
Threadwise and temporal state abstraction
Hierarchically Optimal Yet Local (HOLY) Q-learning
Hierarchically Optimal Recursively Decomposed (HORD) Q-learning
Matthias Hölzl 14
Planned Contents
Introduction to CALisp/Poem
Simple TD-learning: bandits
Flat reinforcement learning: navigation
Hierarchical reinforcement learning:
collecting items individually
Threadwise decomposition for hierarchical
reinforcement learning: learning collaboratively
Matthias Hölzl 15
n-armed Bandits
S
search / 1.0 / N(0.1, 1.0)
coll-known / 1.0 / N(0.3, 3.0)
Choice between n actions
Reward depends probabilisticly on
the action choice
No long-term consequences
Simplest form of TD-learning
Matthias Hölzl 16
Flat Learning
XXXXXXXXXXXXXXXXXXXXXXXX Target: (0 0)
XXTT XX XX
XX XXXXXXXXXX XXXX Choices: (N E S W)
XXXXXX XX XX XX Q-values:
XX XX XXXXXXXX XX #((N (Q -1.8))
XX XXXXXX XX XX (E (Q -1.8))
XX XX XXXXXXXX (S (Q -2.25))
XXXXXXXXXX XX XX (W (Q 2.76)))
XX XX XXXX XX Recommended choice is W
XX XX XX XX XX XX
XX XX RR XX
XXXXXXXXXXXXXXXXXXXXXXXX
Matthias Hölzl 17
Flat Learning
(defun simple-robot ()
(call (nav (target-loc (robot-env)))))
(defun nav (loc)
(until (equal (robot-loc) loc)
(with-choice navigate-choice (dir ’(N E S W))
(action navigate-move dir))))
Matthias Hölzl 18
Hierarchical Learning
(defun waste-removal ()
(loop
(choose choose-waste-removal-action
(action sleep ’SLEEP)
(call (pickup-waste))
(call (drop-waste)))))
(defun pickup-waste ()
(call (nav (waste-source)))
(action pickup-waste ’PICKUP))
Matthias Hölzl 19
Multi-Robot Learning
XXXXXXXXXXXXXXXXXXXXXXXX
XXTT XX RR XX
XX XX XX
XXXXXX XX XX
XX XX RW RR XX
XX XX
XX RR WW XX
XX XX
XX RR XX
XX XX
XX XX
XXXXXXXXXXXXXXXXXXXXXXXX
Matthias Hölzl 20
Thank you!
Any Questions?
matthias.hoelzl@ifi.lmu.de
Matthias Hölzl 21

Más contenido relacionado

Más de FET AWARE project - Self Awareness in Autonomic Systems

Más de FET AWARE project - Self Awareness in Autonomic Systems (20)

Academic Course: 02 Self-organization and emergence in networked systems
Academic Course: 02 Self-organization and emergence in networked systemsAcademic Course: 02 Self-organization and emergence in networked systems
Academic Course: 02 Self-organization and emergence in networked systems
 
Academic Course: 01 Self-awarenesss and Computational Self-awareness
Academic Course: 01 Self-awarenesss and Computational Self-awarenessAcademic Course: 01 Self-awarenesss and Computational Self-awareness
Academic Course: 01 Self-awarenesss and Computational Self-awareness
 
Awareness: Layman Seminar Slides
Awareness: Layman Seminar SlidesAwareness: Layman Seminar Slides
Awareness: Layman Seminar Slides
 
Industry Training: 04 Awareness Applications
Industry Training: 04 Awareness ApplicationsIndustry Training: 04 Awareness Applications
Industry Training: 04 Awareness Applications
 
Industry Training: 03 Awareness Simulation
Industry Training: 03 Awareness SimulationIndustry Training: 03 Awareness Simulation
Industry Training: 03 Awareness Simulation
 
Industry Training: 02 Awareness Properties
Industry Training: 02 Awareness PropertiesIndustry Training: 02 Awareness Properties
Industry Training: 02 Awareness Properties
 
Industry Training: 01 Awareness Overview
Industry Training: 01 Awareness OverviewIndustry Training: 01 Awareness Overview
Industry Training: 01 Awareness Overview
 
Towards Systematically Engineering Ensembles - Martin Wirsing
Towards Systematically Engineering Ensembles - Martin WirsingTowards Systematically Engineering Ensembles - Martin Wirsing
Towards Systematically Engineering Ensembles - Martin Wirsing
 
Capturing the Immune System: From the wet-­lab to the robot, building better ...
Capturing the Immune System: From the wet-­lab to the robot, building better ...Capturing the Immune System: From the wet-­lab to the robot, building better ...
Capturing the Immune System: From the wet-­lab to the robot, building better ...
 
Underwater search and rescue in swarm robotics - Mark Read
Underwater search and rescue in swarm robotics - Mark Read Underwater search and rescue in swarm robotics - Mark Read
Underwater search and rescue in swarm robotics - Mark Read
 
Computational Self-awareness in Smart-Camera Networks - Lukas Esterle
Computational Self-awareness in Smart-Camera Networks - Lukas EsterleComputational Self-awareness in Smart-Camera Networks - Lukas Esterle
Computational Self-awareness in Smart-Camera Networks - Lukas Esterle
 
Why Robots may need to be self-­‐aware, before we can really trust them - Ala...
Why Robots may need to be self-­‐aware, before we can really trust them - Ala...Why Robots may need to be self-­‐aware, before we can really trust them - Ala...
Why Robots may need to be self-­‐aware, before we can really trust them - Ala...
 
Morphogenetic Engineering: Reconciling Architecture and Self-Organization Thr...
Morphogenetic Engineering: Reconciling Architecture and Self-Organization Thr...Morphogenetic Engineering: Reconciling Architecture and Self-Organization Thr...
Morphogenetic Engineering: Reconciling Architecture and Self-Organization Thr...
 
Ensemble-oriented programming of self-adaptive systems - Michele Loreti
Ensemble-oriented programming of self-adaptive systems - Michele LoretiEnsemble-oriented programming of self-adaptive systems - Michele Loreti
Ensemble-oriented programming of self-adaptive systems - Michele Loreti
 
Self-awareness and Adaptive Technologies: the Future of Operating Systems?
Self-awareness and Adaptive Technologies: the Future of Operating Systems? Self-awareness and Adaptive Technologies: the Future of Operating Systems?
Self-awareness and Adaptive Technologies: the Future of Operating Systems?
 
EnhancingWeb Process Self-Awareness with Context-Aware Service Composition
EnhancingWeb Process Self-Awareness with Context-Aware Service CompositionEnhancingWeb Process Self-Awareness with Context-Aware Service Composition
EnhancingWeb Process Self-Awareness with Context-Aware Service Composition
 
Testing cooperative autonomous systems for unwanted emergent behaviour and da...
Testing cooperative autonomous systems for unwanted emergent behaviour and da...Testing cooperative autonomous systems for unwanted emergent behaviour and da...
Testing cooperative autonomous systems for unwanted emergent behaviour and da...
 
Enduring Institutions and Self-Organising Trust-Adaptive Systems for an Open ...
Enduring Institutions and Self-Organising Trust-Adaptive Systems for an Open ...Enduring Institutions and Self-Organising Trust-Adaptive Systems for an Open ...
Enduring Institutions and Self-Organising Trust-Adaptive Systems for an Open ...
 
SmartContent: A self protecting and context aware active content
SmartContent: A self protecting and context aware active contentSmartContent: A self protecting and context aware active content
SmartContent: A self protecting and context aware active content
 
Autonomic Systems Research
Autonomic Systems ResearchAutonomic Systems Research
Autonomic Systems Research
 

Último

Event-Driven Architecture Masterclass: Engineering a Robust, High-performance...
Event-Driven Architecture Masterclass: Engineering a Robust, High-performance...Event-Driven Architecture Masterclass: Engineering a Robust, High-performance...
Event-Driven Architecture Masterclass: Engineering a Robust, High-performance...ScyllaDB
 
Using IESVE for Room Loads Analysis - UK & Ireland
Using IESVE for Room Loads Analysis - UK & IrelandUsing IESVE for Room Loads Analysis - UK & Ireland
Using IESVE for Room Loads Analysis - UK & IrelandIES VE
 
Design Guidelines for Passkeys 2024.pptx
Design Guidelines for Passkeys 2024.pptxDesign Guidelines for Passkeys 2024.pptx
Design Guidelines for Passkeys 2024.pptxFIDO Alliance
 
Linux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdf
Linux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdfLinux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdf
Linux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdfFIDO Alliance
 
Introduction to FIDO Authentication and Passkeys.pptx
Introduction to FIDO Authentication and Passkeys.pptxIntroduction to FIDO Authentication and Passkeys.pptx
Introduction to FIDO Authentication and Passkeys.pptxFIDO Alliance
 
Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...
Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...
Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...panagenda
 
ERP Contender Series: Acumatica vs. Sage Intacct
ERP Contender Series: Acumatica vs. Sage IntacctERP Contender Series: Acumatica vs. Sage Intacct
ERP Contender Series: Acumatica vs. Sage IntacctBrainSell Technologies
 
How we scaled to 80K users by doing nothing!.pdf
How we scaled to 80K users by doing nothing!.pdfHow we scaled to 80K users by doing nothing!.pdf
How we scaled to 80K users by doing nothing!.pdfSrushith Repakula
 
AI mind or machine power point presentation
AI mind or machine power point presentationAI mind or machine power point presentation
AI mind or machine power point presentationyogeshlabana357357
 
Intro in Product Management - Коротко про професію продакт менеджера
Intro in Product Management - Коротко про професію продакт менеджераIntro in Product Management - Коротко про професію продакт менеджера
Intro in Product Management - Коротко про професію продакт менеджераMark Opanasiuk
 
Collecting & Temporal Analysis of Behavioral Web Data - Tales From The Inside
Collecting & Temporal Analysis of Behavioral Web Data - Tales From The InsideCollecting & Temporal Analysis of Behavioral Web Data - Tales From The Inside
Collecting & Temporal Analysis of Behavioral Web Data - Tales From The InsideStefan Dietze
 
How Red Hat Uses FDO in Device Lifecycle _ Costin and Vitaliy at Red Hat.pdf
How Red Hat Uses FDO in Device Lifecycle _ Costin and Vitaliy at Red Hat.pdfHow Red Hat Uses FDO in Device Lifecycle _ Costin and Vitaliy at Red Hat.pdf
How Red Hat Uses FDO in Device Lifecycle _ Costin and Vitaliy at Red Hat.pdfFIDO Alliance
 
TEST BANK For, Information Technology Project Management 9th Edition Kathy Sc...
TEST BANK For, Information Technology Project Management 9th Edition Kathy Sc...TEST BANK For, Information Technology Project Management 9th Edition Kathy Sc...
TEST BANK For, Information Technology Project Management 9th Edition Kathy Sc...marcuskenyatta275
 
2024 May Patch Tuesday
2024 May Patch Tuesday2024 May Patch Tuesday
2024 May Patch TuesdayIvanti
 
Breaking Down the Flutterwave Scandal What You Need to Know.pdf
Breaking Down the Flutterwave Scandal What You Need to Know.pdfBreaking Down the Flutterwave Scandal What You Need to Know.pdf
Breaking Down the Flutterwave Scandal What You Need to Know.pdfUK Journal
 
Introduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdf
Introduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdfIntroduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdf
Introduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdfFIDO Alliance
 
WebRTC and SIP not just audio and video @ OpenSIPS 2024
WebRTC and SIP not just audio and video @ OpenSIPS 2024WebRTC and SIP not just audio and video @ OpenSIPS 2024
WebRTC and SIP not just audio and video @ OpenSIPS 2024Lorenzo Miniero
 
(Explainable) Data-Centric AI: what are you explaininhg, and to whom?
(Explainable) Data-Centric AI: what are you explaininhg, and to whom?(Explainable) Data-Centric AI: what are you explaininhg, and to whom?
(Explainable) Data-Centric AI: what are you explaininhg, and to whom?Paolo Missier
 
ADP Passwordless Journey Case Study.pptx
ADP Passwordless Journey Case Study.pptxADP Passwordless Journey Case Study.pptx
ADP Passwordless Journey Case Study.pptxFIDO Alliance
 

Último (20)

Event-Driven Architecture Masterclass: Engineering a Robust, High-performance...
Event-Driven Architecture Masterclass: Engineering a Robust, High-performance...Event-Driven Architecture Masterclass: Engineering a Robust, High-performance...
Event-Driven Architecture Masterclass: Engineering a Robust, High-performance...
 
Using IESVE for Room Loads Analysis - UK & Ireland
Using IESVE for Room Loads Analysis - UK & IrelandUsing IESVE for Room Loads Analysis - UK & Ireland
Using IESVE for Room Loads Analysis - UK & Ireland
 
Design Guidelines for Passkeys 2024.pptx
Design Guidelines for Passkeys 2024.pptxDesign Guidelines for Passkeys 2024.pptx
Design Guidelines for Passkeys 2024.pptx
 
Linux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdf
Linux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdfLinux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdf
Linux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdf
 
Introduction to FIDO Authentication and Passkeys.pptx
Introduction to FIDO Authentication and Passkeys.pptxIntroduction to FIDO Authentication and Passkeys.pptx
Introduction to FIDO Authentication and Passkeys.pptx
 
Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...
Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...
Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...
 
ERP Contender Series: Acumatica vs. Sage Intacct
ERP Contender Series: Acumatica vs. Sage IntacctERP Contender Series: Acumatica vs. Sage Intacct
ERP Contender Series: Acumatica vs. Sage Intacct
 
How we scaled to 80K users by doing nothing!.pdf
How we scaled to 80K users by doing nothing!.pdfHow we scaled to 80K users by doing nothing!.pdf
How we scaled to 80K users by doing nothing!.pdf
 
AI mind or machine power point presentation
AI mind or machine power point presentationAI mind or machine power point presentation
AI mind or machine power point presentation
 
Intro in Product Management - Коротко про професію продакт менеджера
Intro in Product Management - Коротко про професію продакт менеджераIntro in Product Management - Коротко про професію продакт менеджера
Intro in Product Management - Коротко про професію продакт менеджера
 
Collecting & Temporal Analysis of Behavioral Web Data - Tales From The Inside
Collecting & Temporal Analysis of Behavioral Web Data - Tales From The InsideCollecting & Temporal Analysis of Behavioral Web Data - Tales From The Inside
Collecting & Temporal Analysis of Behavioral Web Data - Tales From The Inside
 
How Red Hat Uses FDO in Device Lifecycle _ Costin and Vitaliy at Red Hat.pdf
How Red Hat Uses FDO in Device Lifecycle _ Costin and Vitaliy at Red Hat.pdfHow Red Hat Uses FDO in Device Lifecycle _ Costin and Vitaliy at Red Hat.pdf
How Red Hat Uses FDO in Device Lifecycle _ Costin and Vitaliy at Red Hat.pdf
 
Overview of Hyperledger Foundation
Overview of Hyperledger FoundationOverview of Hyperledger Foundation
Overview of Hyperledger Foundation
 
TEST BANK For, Information Technology Project Management 9th Edition Kathy Sc...
TEST BANK For, Information Technology Project Management 9th Edition Kathy Sc...TEST BANK For, Information Technology Project Management 9th Edition Kathy Sc...
TEST BANK For, Information Technology Project Management 9th Edition Kathy Sc...
 
2024 May Patch Tuesday
2024 May Patch Tuesday2024 May Patch Tuesday
2024 May Patch Tuesday
 
Breaking Down the Flutterwave Scandal What You Need to Know.pdf
Breaking Down the Flutterwave Scandal What You Need to Know.pdfBreaking Down the Flutterwave Scandal What You Need to Know.pdf
Breaking Down the Flutterwave Scandal What You Need to Know.pdf
 
Introduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdf
Introduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdfIntroduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdf
Introduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdf
 
WebRTC and SIP not just audio and video @ OpenSIPS 2024
WebRTC and SIP not just audio and video @ OpenSIPS 2024WebRTC and SIP not just audio and video @ OpenSIPS 2024
WebRTC and SIP not just audio and video @ OpenSIPS 2024
 
(Explainable) Data-Centric AI: what are you explaininhg, and to whom?
(Explainable) Data-Centric AI: what are you explaininhg, and to whom?(Explainable) Data-Centric AI: what are you explaininhg, and to whom?
(Explainable) Data-Centric AI: what are you explaininhg, and to whom?
 
ADP Passwordless Journey Case Study.pptx
ADP Passwordless Journey Case Study.pptxADP Passwordless Journey Case Study.pptx
ADP Passwordless Journey Case Study.pptx
 

Robot Swarms as Ensembles of Cooperating Components - Matthias Holzl

  • 1. Robot Swarms as Ensembles of Cooperating Components Matthias Hölzl With contributions from Martin Wirsing, Annabelle Klarl AWASS Lucca, June 24, 2013 www.ascens-ist.eu
  • 2. The Task Robots cleaning an exhibition area Matthias Hölzl 2
  • 3. marXbot Miniature mobile robot developed by EPFL Rough-terrain mobility Robots can dock to other robots Many sensors Proximity sensors Gyroscope 3D accelerometer RFID reader Cameras . . . ARM-based Linux system Gripper for picking up items Matthias Hölzl 3
  • 5. Problems Noise, sensor resolution Extracting information from sensor data Unforeseen situations Uncertainty about the environment Performing complex actions when intermediate results are uncertain . . . Matthias Hölzl 5
  • 6. Action Logics Logics that can represent change over time Probabilistic behavior can be modeled (but is cumbersome) Matthias Hölzl 6
  • 7. Markov Decision Processes pos = (x,y) pos = (x,y+1) pos = (x+1,y+1) pos = (x + 1,y) e / ... s, n / ... s / 0.9 / -0.1 e,w / 0.025 / -0.1 w / ... s,n / ... e / ... s, n / ... w/ ... s,n / ... n / 0.9 / -0.1 e, w / 0.025 / -0.1 s,n,w,e / 0.05 / -0.1 Matthias Hölzl 7
  • 8. Markov Decision Processes watchTV goToClub Decide Activity In Club Watching TV Dancing Alone Drinking Dancing With Partner dance p = 0.5 dance p = 0.5 drinkBeer Oh, oh Oh, no flirt p = 0.05 p = 0.95 Matthias Hölzl 8
  • 9. MDPs: Strategies watchTV goToClub Decide Activity In Club Watching TV Dancing Alone Drinking Dancing With Partner dance p = 0.5 dance p = 0.5 drinkBeer Oh, oh Oh, no flirt p = 0.05 p = 0.95 State TV CDB CDF DA watchTV goToClub goToClub IC drinkBeer drinkBeer dance DWP flirt flirt flirt Utility 0.1 −0.05(∗) −1.975(∗) (∗) 0.05 + (−0.1) (∗∗) 0.05 + (0.5 × 0.2) + 0.5 × (0.25 + (0.05 × 5) + (0.95 × −5)) Matthias Hölzl 9
  • 10. Reinforcement Learning General idea: Figure out the expected value of each action in each state Pick the action with the highest expected value (most of the time) Update the expectations according to the actual rewards Matthias Hölzl 10
  • 11. How well does this work? Rather well for small problems But: state explosion Matthias Hölzl 11
  • 13. POEM Action language First-order reasoning Hierarchical reinforcement learning Learns completions for partial programs Concurrency Reflection / meta-object protocols · · · Matthias Hölzl 13
  • 14. Iliad: A POEM Implementation Common Lisp-based programming language Full first-order reasoning Operations on logical theories: U(C)NA, domain closure, . . . Resolution, hyperresolution, DPLL, etc. Conditional answer extraction Procedural attachment, constraint solving Hierarchical reinforcement learning Based on Concurrent ALisp Partial programs Threadwise and temporal state abstraction Hierarchically Optimal Yet Local (HOLY) Q-learning Hierarchically Optimal Recursively Decomposed (HORD) Q-learning Matthias Hölzl 14
  • 15. Planned Contents Introduction to CALisp/Poem Simple TD-learning: bandits Flat reinforcement learning: navigation Hierarchical reinforcement learning: collecting items individually Threadwise decomposition for hierarchical reinforcement learning: learning collaboratively Matthias Hölzl 15
  • 16. n-armed Bandits S search / 1.0 / N(0.1, 1.0) coll-known / 1.0 / N(0.3, 3.0) Choice between n actions Reward depends probabilisticly on the action choice No long-term consequences Simplest form of TD-learning Matthias Hölzl 16
  • 17. Flat Learning XXXXXXXXXXXXXXXXXXXXXXXX Target: (0 0) XXTT XX XX XX XXXXXXXXXX XXXX Choices: (N E S W) XXXXXX XX XX XX Q-values: XX XX XXXXXXXX XX #((N (Q -1.8)) XX XXXXXX XX XX (E (Q -1.8)) XX XX XXXXXXXX (S (Q -2.25)) XXXXXXXXXX XX XX (W (Q 2.76))) XX XX XXXX XX Recommended choice is W XX XX XX XX XX XX XX XX RR XX XXXXXXXXXXXXXXXXXXXXXXXX Matthias Hölzl 17
  • 18. Flat Learning (defun simple-robot () (call (nav (target-loc (robot-env))))) (defun nav (loc) (until (equal (robot-loc) loc) (with-choice navigate-choice (dir ’(N E S W)) (action navigate-move dir)))) Matthias Hölzl 18
  • 19. Hierarchical Learning (defun waste-removal () (loop (choose choose-waste-removal-action (action sleep ’SLEEP) (call (pickup-waste)) (call (drop-waste))))) (defun pickup-waste () (call (nav (waste-source))) (action pickup-waste ’PICKUP)) Matthias Hölzl 19
  • 20. Multi-Robot Learning XXXXXXXXXXXXXXXXXXXXXXXX XXTT XX RR XX XX XX XX XXXXXX XX XX XX XX RW RR XX XX XX XX RR WW XX XX XX XX RR XX XX XX XX XX XXXXXXXXXXXXXXXXXXXXXXXX Matthias Hölzl 20