SlideShare a Scribd company logo
1 of 62
Download to read offline
optimization of probabilistic argumentation
with markov processes
E. Hadoux1
, A. Beynier1
, N. Maudet1
, P. Weng2
and A. Hunter3
Tue., Sept. 29th
(1) Sorbonne Universités, UPMC Univ Paris 6, UMR 7606, LIP6, F-75005, Paris, France
(2) SYSU-CMU Joint Institute of Engineering, Guangzhou, China
SYSU-CMU Shunde International Joint Research Institute, Shunde, China
(3) Department of Computer Science, University College London, Gower Street, London
WC1E 6BT, UK
Introduction
∙ Debate argumentation problems between two agents
1
Introduction
∙ Debate argumentation problems between two agents
∙ Probabilistic executable logic to improve expressivity
1
Introduction
∙ Debate argumentation problems between two agents
∙ Probabilistic executable logic to improve expressivity
∙ New class of problems: Argumentation Problem with
Probabilistic Strategies (APS) (Hunter, 2014)
1
Introduction
∙ Debate argumentation problems between two agents
∙ Probabilistic executable logic to improve expressivity
∙ New class of problems: Argumentation Problem with
Probabilistic Strategies (APS) (Hunter, 2014)
∙ Purpose of this work: optimize the sequence of arguments
of one agent
1
Introduction
∙ Debate argumentation problems between two agents
∙ Probabilistic executable logic to improve expressivity
∙ New class of problems: Argumentation Problem with
Probabilistic Strategies (APS) (Hunter, 2014)
∙ Purpose of this work: optimize the sequence of arguments
of one agent
There will be abuse of the word predicate!
1
formalization
Formalization of a debate problem
∙ Turn-based game between two agents
∙ Rules to fire in order to attack arguments of the opponent
and revise knowledge
3
Formalization of a debate problem
∙ Turn-based game between two agents
∙ Rules to fire in order to attack arguments of the opponent
and revise knowledge
Let us define a debate problem with:
∙ A, the set or arguments
3
Formalization of a debate problem
∙ Turn-based game between two agents
∙ Rules to fire in order to attack arguments of the opponent
and revise knowledge
Let us define a debate problem with:
∙ A, the set or arguments
∙ E, the set of attacks
3
Formalization of a debate problem
∙ Turn-based game between two agents
∙ Rules to fire in order to attack arguments of the opponent
and revise knowledge
Let us define a debate problem with:
∙ A, the set or arguments
∙ E, the set of attacks
∙ P = 2A × 2E, the public space gathering voiced arguments
3
Formalization of a debate problem
∙ Turn-based game between two agents
∙ Rules to fire in order to attack arguments of the opponent
and revise knowledge
Let us define a debate problem with:
∙ A, the set or arguments
∙ E, the set of attacks
∙ P = 2A × 2E, the public space gathering voiced arguments
∙ Two agents: agent 1 and agent 2
3
Notation
∙ Arguments: literals (e.g., a, b, c)
4
Notation
∙ Arguments: literals (e.g., a, b, c)
∙ Attacks: e(x, y) if x attacks y
4
Notation
∙ Arguments: literals (e.g., a, b, c)
∙ Attacks: e(x, y) if x attacks y
∙ Args. in public (resp. private) space: a(x) (resp. hi(x))
4
Notation
∙ Arguments: literals (e.g., a, b, c)
∙ Attacks: e(x, y) if x attacks y
∙ Args. in public (resp. private) space: a(x) (resp. hi(x))
∙ Goals:
∧
k g(xk) (resp. g(¬xk)) if xk is (resp. is not) accepted
in the public space (Dung, 1995)
4
Notation
∙ Arguments: literals (e.g., a, b, c)
∙ Attacks: e(x, y) if x attacks y
∙ Args. in public (resp. private) space: a(x) (resp. hi(x))
∙ Goals:
∧
k g(xk) (resp. g(¬xk)) if xk is (resp. is not) accepted
in the public space (Dung, 1995)
∙ Rules: prem ⇒ Pr(Acts)
4
Notation
∙ Arguments: literals (e.g., a, b, c)
∙ Attacks: e(x, y) if x attacks y
∙ Args. in public (resp. private) space: a(x) (resp. hi(x))
∙ Goals:
∧
k g(xk) (resp. g(¬xk)) if xk is (resp. is not) accepted
in the public space (Dung, 1995)
∙ Rules: prem ⇒ Pr(Acts)
∙ Premises: conjunctions of e(, ), a(), hi()
4
Notation
∙ Arguments: literals (e.g., a, b, c)
∙ Attacks: e(x, y) if x attacks y
∙ Args. in public (resp. private) space: a(x) (resp. hi(x))
∙ Goals:
∧
k g(xk) (resp. g(¬xk)) if xk is (resp. is not) accepted
in the public space (Dung, 1995)
∙ Rules: prem ⇒ Pr(Acts)
∙ Premises: conjunctions of e(, ), a(), hi()
∙ Acts: conjunctions of ⊞, ⊟ on e(, ), a() and ⊕, ⊖ on hi()
4
Formalization of an APS
An APS is characterized (from the point of view of agent 1) by
⟨A, E, G, S1, g1, g2, S2, P, R1, R2⟩:
∙ A, E, P as specified above
∙ G, the set of all possible goals
∙ Si, the set of private states for agent i
∙ gi ∈ G, the given goal for agent i
∙ Ri, the set of rules for agent i
5
Example: Arguments
Is e-sport a sport?
6
Example: Arguments
Is e-sport a sport?
a E-sport is a sport
b E-sport requires focusing,
precision and generates
tiredness
c Not all sports are physical
d Sports not referenced by IOC
exist
e Chess is a sport
f E-sport is not a physical
activity
g E-sport is not referenced by
IOC
h Working requires focusing and
generates tiredness but is not
a sport
6
Example: Formalization
∙ A = {a, b, c, d, e, f, g, h}
7
Example: Formalization
∙ A = {a, b, c, d, e, f, g, h}
∙ E = { e(f, a), e(g, a), e(b, f), e(c, f), e(h, b), e(g, c),
e(d, g), e(e, g)}
7
Example: Formalization
∙ A = {a, b, c, d, e, f, g, h}
∙ E = { e(f, a), e(g, a), e(b, f), e(c, f), e(h, b), e(g, c),
e(d, g), e(e, g)}
∙ g1 = g(a)
7
Example: Formalization
∙ A = {a, b, c, d, e, f, g, h}
∙ E = { e(f, a), e(g, a), e(b, f), e(c, f), e(h, b), e(g, c),
e(d, g), e(e, g)}
∙ g1 = g(a)
∙ R1 = {h1(a) ⇒ ⊞a(a),
h1(b) ∧ a(f) ∧ h1(c) ∧ e(b, f) ∧ e(c, f) ⇒
0.5 : ⊞a(b) ∧ ⊞e(b, f) ∨ 0.5 : ⊞a(c) ∧ ⊞e(c, f),
h1(d) ∧ a(g) ∧ h1(e) ∧ e(d, g) ∧ e(e, g) ⇒
0.8 : ⊞a(e) ∧ ⊞e(e, g) ∨ 0.2 : ⊞a(d) ∧ ⊞e(d, g)}
7
Example: Formalization
∙ R2 = {h2(h) ∧ a(b) ∧ e(h, b) ⇒ ⊞a(h) ∧ ⊞e(h, b),
h2(g) ∧ a(c) ∧ e(g, c) ⇒ ⊞a(g) ∧ ⊞e(g, c),
a(a) ∧ h2(f) ∧ h2(g) ∧ e(f, a) ⇒
0.8 : ⊞a(f) ∧ ⊞e(f, a) ∨ 0.2 : ⊞a(g) ∧ ⊞e(g, a)}
∙ Initial state: h1(a, b, c, d, e), {}, h2(f, g, h)
8
Attacks graph
a
g f
c bde
h
Figure: Graph of arguments of Example e-sport
9
Probabilistic Finite State Machine: Graph
APS → Probabilistic Finite State Machine
σ1start σ2
σ3
σ4
σ5
σ6 σ7
σ8 σ9 σ10
σ11
σ12
1
0.8
0.2
0.5
0.5 1
1
0.8 0.2
0.8
0.2
Figure: PFSM of Example e-sport
10
Probabilistic Finite State Machine
To optimize the sequence of arguments for agent 1, we could
optimize the PFSM but:
11
Probabilistic Finite State Machine
To optimize the sequence of arguments for agent 1, we could
optimize the PFSM but:
1. depends of the initial state
11
Probabilistic Finite State Machine
To optimize the sequence of arguments for agent 1, we could
optimize the PFSM but:
1. depends of the initial state
2. requires knowledge of the private state of the opponent
11
Probabilistic Finite State Machine
To optimize the sequence of arguments for agent 1, we could
optimize the PFSM but:
1. depends of the initial state
2. requires knowledge of the private state of the opponent
Using Markov models, we can relax assumptions 1 and 2.
Moreover, the APS formalization can be modified in order to
comply with the Markov assumption.
11
Markov Decision Process
A Markov Decision Process (MDP) (Puterman, 1994) is
characterized by a tuple ⟨S, A, T, R⟩:
∙ S, a set of states,
∙ A, a set of actions,
∙ T : S × A → Pr(S), a transition function,
∙ R : S × A → R, a reward function.
12
Partially-Observable Markov Decision Process
A Partially-Observable MDP (POMDP) (Puterman, 1994) is
characterized by a tuple ⟨S, A, T, R, O, Q⟩:
∙ S, a set of states,
∙ A, a set of actions,
∙ T : S × A → Pr(S), a transition function,
∙ R : S × A → R, a reward function,
∙ O, an observation set,
∙ Q : S × A → Pr(O), an observation function.
13
Mixed-Observability Markov Decision Process
A Mixed-Observability MDP (MOMDP) (Ong et al., 2010) is
characterized by a tuple ⟨Sv, Sh, A, T, R, Ov, Oh, Q⟩:
∙ Sv, Sh, a visible and hidden parts of the state,
∙ A, a set of actions,
∙ T : Sv × A × Sh → Pr(Sv × Sh), a transition function,
∙ R : Sv × A × Sh → R, a reward function,
∙ Ov = Sv, an observation set on the visible part of the state,
∙ Oh, an observation set on the hidden part of the state,
∙ Q : Sv × A × Sh → Pr(Ov × Oh), an observation function.
14
transformation to a momdp
Transformation to a MOMDP
An APS from the point of view of agent 1 can be transformed to
a MOMDP:
∙ Sv = S1 × P, Sh = S2
∙ A = {prem(r) ⇒ m|r ∈ R1 and m ∈ acts(r)}
∙ Ov = Sv and Oh = ∅
∙ Q(⟨sv, sh⟩, a, ⟨sv⟩) = 1, otherwise 0
∙ T, see after
16
Transformation to a MOMDP: Transition function
Application set
Let Cs(Ri) be the set of rules of Ri that can be fired in state s.
The application set Fr(m, s) is the set of predicates resulting
from the application of act m of a rule r on s. If r cannot be
fired in s, Fr(m, s) = s.
∙ s, a state and r : p ⇒ m, an action s.t. r ∈ A
∙ s′ = Fr(m, s)
∙ r′ ∈ Cs′ (R2) s.t. r′ : p′ ⇒ [π1/m1, . . . , πn/mn]
∙ s′′
i = Fr′ (mi, s′)
∙ T(s, r, s′′
i ) = πi
17
Reward function
For the reward function:
∙ with Dung’s semantics: positive reward for each part holding
∙ can be generalized: General Gradual Valuation (Cayrol and
Lagasquie-Schiex, 2005)
18
Transformation to a MOMDP
Model sizes:
APS : 8 arguments, 8 attacks, 6 rules
POMDP : 4 294 967 296 states
MOMDP : 16 777 216 states
Untractable instances → need to optimize at the root
19
solving an aps
Solving an APS
Two algorithms to solve MOMDPs:
∙ MO-IP (Araya-López et al., 2010), IP of POMDP on MOMDP
(exact method)
∙ MO-SARSOP (Ong et al., 2010), SARSOP of POMDP on MOMDP
(approximate method albeit very efficient)
Two kinds of optimizations: with or without dependencies on
the initial state
21
Optimizations without dependencies
Irr. Prunes irrelevant arguments
22
Optimizations without dependencies
Irr. Prunes irrelevant arguments
Enth. Infers attacks
22
Optimizations without dependencies
Irr. Prunes irrelevant arguments
Enth. Infers attacks
Dom. Removes dominated arguments
22
Optimizations without dependencies
Irr. Prunes irrelevant arguments
Enth. Infers attacks
Dom. Removes dominated arguments
Guarantee on the unicity and optimality of the solution.
22
Attacks graph
Argument dominance
If an argument is attacked by
any unattacked argument, it is
dominated.
a f
g
b
c
d e
h
Figure: Attacks graph of Example
23
Optimization with dependencies
Irr(s0) has to be reapplied each time the initial state changes.
24
Optimization with dependencies
Irr(s0) has to be reapplied each time the initial state changes.
1. For each predicate that is never modified but used as
premises:
1.1 Remove all the rules that are not compatible with the value of
this predicate in the initial state.
1.2 For all remaining rules, remove the predicate from the premises.
24
Optimization with dependencies
Irr(s0) has to be reapplied each time the initial state changes.
1. For each predicate that is never modified but used as
premises:
1.1 Remove all the rules that are not compatible with the value of
this predicate in the initial state.
1.2 For all remaining rules, remove the predicate from the premises.
2. For each remaining action of agent 1, track the rules of agent
2 compatible with the application of this action. If a rule of
agent 2 is not compatible with any application of an action
of agent 1, remove it.
24
experiments
Experiments
We computed a solution for the e-sport problem with:
∙ MO-IP, which did not finish after tens of hours
∙ MO-SARSOP without optimizations, idem
∙ MO-SARSOP with optimizations, 4sec for the optimal solution
26
Experiments: Policy graph
r1
1,1start r1
2,2 r1
3,1 ∅
r1
3,1∅ r1
2,2
∅ ∅
o2 o5o4
o6
o7
o8
o5
o1
o7
o8
o3
o3
o4
Figure: Policy graph for Example
27
Experiments: More examples
None Irr. Enth. Dom. Irr(s0). All
Ex 1 — — — — — 0.56
Ex 2 3.3 0.3 0.3 0.4 0 0
Dv. — — — — — 32
6 1313 22 43 7 2.4 0.9
7 — 180 392 16 20 6.7
8 — — — — 319 45
9 — — — — — —
Table: Computation time (in seconds)
28
conclusion and discussions
Conclusion
We presented:
1. A new framework to represent more complex debate
problems (APS)
2. A method to transform those problems to a MOMDP
3. Several optimizations that can be used outside of the
context of MOMDP
4. A method to optimize actions of an agent in an APS
30
Perspectives
We are currently working on using POMCP (Silver and Veness,
2010).
We are also using HS3MDPs (Hadoux et al., 2014).
31
Questions?
32
Bibliography I
Araya-López, M., Thomas, V., Buffet, O., and Charpillet, F. (2010).
A closer look at MOMDPs. In 22nd IEEE International
Conference on Tools with Artificial Intelligence (ICTAI).
Cayrol, C. and Lagasquie-Schiex, M.-C. (2005). Graduality in
argumentation. Journal of Artificial Intelligence Research
(JAIR), 23:245–297.
Dung, P. M. (1995). On the acceptability of arguments and its
fundamental role in nonmonotonic reasoning, logic
programming and n-person games. Artificial Intelligence,
77(2):321–358.
33
Bibliography II
Hadoux, E., Beynier, A., and Weng, P. (2014). Solving
Hidden-Semi-Markov-Mode Markov Decision Problems. In
Straccia, U. and Calì, A., editors, Scalable Uncertainty
Management, volume 8720 of Lecture Notes in Computer
Science, pages 176–189. Springer International Publishing.
Hunter, A. (2014). Probabilistic strategies in dialogical
argumentation. In International Conference on Scalable
Uncertainty Management (SUM’14) LNCS volume 8720.
Ong, S. C., Png, S. W., Hsu, D., and Lee, W. S. (2010). Planning
under uncertainty for robotic tasks with mixed observability.
In The International Journal of Robotics Research.
34
Bibliography III
Puterman, M. L. (1994). Markov Decision Processes: discrete
stochastic dynamic programming. John Wiley & Sons.
Silver, D. and Veness, J. (2010). Monte-Carlo planning in large
POMDPs. In Proceedings of the 24th Conference on Neural
Information Processing Systems (NIPS), pages 2164–2172.
35

More Related Content

What's hot

29 conservative fields potential functions
29 conservative fields potential functions29 conservative fields potential functions
29 conservative fields potential functions
math267
 

What's hot (19)

On the Jensen-Shannon symmetrization of distances relying on abstract means
On the Jensen-Shannon symmetrization of distances relying on abstract meansOn the Jensen-Shannon symmetrization of distances relying on abstract means
On the Jensen-Shannon symmetrization of distances relying on abstract means
 
Side 2019 #5
Side 2019 #5Side 2019 #5
Side 2019 #5
 
Slides risk-rennes
Slides risk-rennesSlides risk-rennes
Slides risk-rennes
 
Slides ensae 8
Slides ensae 8Slides ensae 8
Slides ensae 8
 
Slides simplexe
Slides simplexeSlides simplexe
Slides simplexe
 
Sildes buenos aires
Sildes buenos airesSildes buenos aires
Sildes buenos aires
 
Slides econ-lm
Slides econ-lmSlides econ-lm
Slides econ-lm
 
transformations and nonparametric inference
transformations and nonparametric inferencetransformations and nonparametric inference
transformations and nonparametric inference
 
Quantile and Expectile Regression
Quantile and Expectile RegressionQuantile and Expectile Regression
Quantile and Expectile Regression
 
Proba stats-r1-2017
Proba stats-r1-2017Proba stats-r1-2017
Proba stats-r1-2017
 
Slides amsterdam-2013
Slides amsterdam-2013Slides amsterdam-2013
Slides amsterdam-2013
 
Slides toulouse
Slides toulouseSlides toulouse
Slides toulouse
 
29 conservative fields potential functions
29 conservative fields potential functions29 conservative fields potential functions
29 conservative fields potential functions
 
Inequality #4
Inequality #4Inequality #4
Inequality #4
 
Slides univ-van-amsterdam
Slides univ-van-amsterdamSlides univ-van-amsterdam
Slides univ-van-amsterdam
 
Quantum optical models in noncommutative spaces
Quantum optical models in noncommutative spacesQuantum optical models in noncommutative spaces
Quantum optical models in noncommutative spaces
 
Hands-On Algorithms for Predictive Modeling
Hands-On Algorithms for Predictive ModelingHands-On Algorithms for Predictive Modeling
Hands-On Algorithms for Predictive Modeling
 
Slides lln-risques
Slides lln-risquesSlides lln-risques
Slides lln-risques
 
2 integration and the substitution methods x
2 integration and the substitution methods x2 integration and the substitution methods x
2 integration and the substitution methods x
 

Similar to Optimization of probabilistic argumentation with Markov processes

Runtime Analysis of Population-based Evolutionary Algorithms
Runtime Analysis of Population-based Evolutionary AlgorithmsRuntime Analysis of Population-based Evolutionary Algorithms
Runtime Analysis of Population-based Evolutionary Algorithms
Per Kristian Lehre
 
Runtime Analysis of Population-based Evolutionary Algorithms
Runtime Analysis of Population-based Evolutionary AlgorithmsRuntime Analysis of Population-based Evolutionary Algorithms
Runtime Analysis of Population-based Evolutionary Algorithms
PK Lehre
 
Poggi analytics - star - 1a
Poggi   analytics - star - 1aPoggi   analytics - star - 1a
Poggi analytics - star - 1a
Gaston Liberman
 
Lecture 3 qualtifed rules of inference
Lecture 3 qualtifed rules of inferenceLecture 3 qualtifed rules of inference
Lecture 3 qualtifed rules of inference
asimnawaz54
 

Similar to Optimization of probabilistic argumentation with Markov processes (20)

Runtime Analysis of Population-based Evolutionary Algorithms
Runtime Analysis of Population-based Evolutionary AlgorithmsRuntime Analysis of Population-based Evolutionary Algorithms
Runtime Analysis of Population-based Evolutionary Algorithms
 
Runtime Analysis of Population-based Evolutionary Algorithms
Runtime Analysis of Population-based Evolutionary AlgorithmsRuntime Analysis of Population-based Evolutionary Algorithms
Runtime Analysis of Population-based Evolutionary Algorithms
 
Compiler Construction | Lecture 9 | Constraint Resolution
Compiler Construction | Lecture 9 | Constraint ResolutionCompiler Construction | Lecture 9 | Constraint Resolution
Compiler Construction | Lecture 9 | Constraint Resolution
 
Declarative Datalog Debugging for Mere Mortals
Declarative Datalog Debugging for Mere MortalsDeclarative Datalog Debugging for Mere Mortals
Declarative Datalog Debugging for Mere Mortals
 
Algorithm Assignment Help
Algorithm Assignment HelpAlgorithm Assignment Help
Algorithm Assignment Help
 
Generic Reinforcement Schemes and Their Optimization
Generic Reinforcement Schemes and Their OptimizationGeneric Reinforcement Schemes and Their Optimization
Generic Reinforcement Schemes and Their Optimization
 
02 math essentials
02 math essentials02 math essentials
02 math essentials
 
C2.0 propositional logic
C2.0 propositional logicC2.0 propositional logic
C2.0 propositional logic
 
Otter 2016-11-28-01-ss
Otter 2016-11-28-01-ssOtter 2016-11-28-01-ss
Otter 2016-11-28-01-ss
 
Hierarchical Reinforcement Learning with Option-Critic Architecture
Hierarchical Reinforcement Learning with Option-Critic ArchitectureHierarchical Reinforcement Learning with Option-Critic Architecture
Hierarchical Reinforcement Learning with Option-Critic Architecture
 
Competitive Multi-agent Inverse Reinforcement Learning with Sub-optimal Demon...
Competitive Multi-agent Inverse Reinforcement Learning with Sub-optimal Demon...Competitive Multi-agent Inverse Reinforcement Learning with Sub-optimal Demon...
Competitive Multi-agent Inverse Reinforcement Learning with Sub-optimal Demon...
 
A New Nonlinear Reinforcement Scheme for Stochastic Learning Automata
A New Nonlinear Reinforcement Scheme for Stochastic Learning AutomataA New Nonlinear Reinforcement Scheme for Stochastic Learning Automata
A New Nonlinear Reinforcement Scheme for Stochastic Learning Automata
 
L03 ai - knowledge representation using logic
L03 ai - knowledge representation using logicL03 ai - knowledge representation using logic
L03 ai - knowledge representation using logic
 
Scala as a Declarative Language
Scala as a Declarative LanguageScala as a Declarative Language
Scala as a Declarative Language
 
Poggi analytics - star - 1a
Poggi   analytics - star - 1aPoggi   analytics - star - 1a
Poggi analytics - star - 1a
 
Cheatsheet supervised-learning
Cheatsheet supervised-learningCheatsheet supervised-learning
Cheatsheet supervised-learning
 
PTSP PPT.pdf
PTSP PPT.pdfPTSP PPT.pdf
PTSP PPT.pdf
 
Side 2019 #7
Side 2019 #7Side 2019 #7
Side 2019 #7
 
Application H-matrices for solving PDEs with multi-scale coefficients, jumpin...
Application H-matrices for solving PDEs with multi-scale coefficients, jumpin...Application H-matrices for solving PDEs with multi-scale coefficients, jumpin...
Application H-matrices for solving PDEs with multi-scale coefficients, jumpin...
 
Lecture 3 qualtifed rules of inference
Lecture 3 qualtifed rules of inferenceLecture 3 qualtifed rules of inference
Lecture 3 qualtifed rules of inference
 

Recently uploaded

CYTOGENETIC MAP................ ppt.pptx
CYTOGENETIC MAP................ ppt.pptxCYTOGENETIC MAP................ ppt.pptx
CYTOGENETIC MAP................ ppt.pptx
Cherry
 
(May 9, 2024) Enhanced Ultrafast Vector Flow Imaging (VFI) Using Multi-Angle ...
(May 9, 2024) Enhanced Ultrafast Vector Flow Imaging (VFI) Using Multi-Angle ...(May 9, 2024) Enhanced Ultrafast Vector Flow Imaging (VFI) Using Multi-Angle ...
(May 9, 2024) Enhanced Ultrafast Vector Flow Imaging (VFI) Using Multi-Angle ...
Scintica Instrumentation
 
POGONATUM : morphology, anatomy, reproduction etc.
POGONATUM : morphology, anatomy, reproduction etc.POGONATUM : morphology, anatomy, reproduction etc.
POGONATUM : morphology, anatomy, reproduction etc.
Cherry
 
Cyathodium bryophyte: morphology, anatomy, reproduction etc.
Cyathodium bryophyte: morphology, anatomy, reproduction etc.Cyathodium bryophyte: morphology, anatomy, reproduction etc.
Cyathodium bryophyte: morphology, anatomy, reproduction etc.
Cherry
 
COMPOSTING : types of compost, merits and demerits
COMPOSTING : types of compost, merits and demeritsCOMPOSTING : types of compost, merits and demerits
COMPOSTING : types of compost, merits and demerits
Cherry
 
Porella : features, morphology, anatomy, reproduction etc.
Porella : features, morphology, anatomy, reproduction etc.Porella : features, morphology, anatomy, reproduction etc.
Porella : features, morphology, anatomy, reproduction etc.
Cherry
 
Lipids: types, structure and important functions.
Lipids: types, structure and important functions.Lipids: types, structure and important functions.
Lipids: types, structure and important functions.
Cherry
 
ONLINE VOTING SYSTEM SE Project for vote
ONLINE VOTING SYSTEM SE Project for voteONLINE VOTING SYSTEM SE Project for vote
ONLINE VOTING SYSTEM SE Project for vote
RaunakRastogi4
 
LUNULARIA -features, morphology, anatomy ,reproduction etc.
LUNULARIA -features, morphology, anatomy ,reproduction etc.LUNULARIA -features, morphology, anatomy ,reproduction etc.
LUNULARIA -features, morphology, anatomy ,reproduction etc.
Cherry
 

Recently uploaded (20)

TransientOffsetin14CAftertheCarringtonEventRecordedbyPolarTreeRings
TransientOffsetin14CAftertheCarringtonEventRecordedbyPolarTreeRingsTransientOffsetin14CAftertheCarringtonEventRecordedbyPolarTreeRings
TransientOffsetin14CAftertheCarringtonEventRecordedbyPolarTreeRings
 
FS P2 COMBO MSTA LAST PUSH past exam papers.
FS P2 COMBO MSTA LAST PUSH past exam papers.FS P2 COMBO MSTA LAST PUSH past exam papers.
FS P2 COMBO MSTA LAST PUSH past exam papers.
 
Role of AI in seed science Predictive modelling and Beyond.pptx
Role of AI in seed science  Predictive modelling and  Beyond.pptxRole of AI in seed science  Predictive modelling and  Beyond.pptx
Role of AI in seed science Predictive modelling and Beyond.pptx
 
CYTOGENETIC MAP................ ppt.pptx
CYTOGENETIC MAP................ ppt.pptxCYTOGENETIC MAP................ ppt.pptx
CYTOGENETIC MAP................ ppt.pptx
 
module for grade 9 for distance learning
module for grade 9 for distance learningmodule for grade 9 for distance learning
module for grade 9 for distance learning
 
(May 9, 2024) Enhanced Ultrafast Vector Flow Imaging (VFI) Using Multi-Angle ...
(May 9, 2024) Enhanced Ultrafast Vector Flow Imaging (VFI) Using Multi-Angle ...(May 9, 2024) Enhanced Ultrafast Vector Flow Imaging (VFI) Using Multi-Angle ...
(May 9, 2024) Enhanced Ultrafast Vector Flow Imaging (VFI) Using Multi-Angle ...
 
Climate Change Impacts on Terrestrial and Aquatic Ecosystems.pptx
Climate Change Impacts on Terrestrial and Aquatic Ecosystems.pptxClimate Change Impacts on Terrestrial and Aquatic Ecosystems.pptx
Climate Change Impacts on Terrestrial and Aquatic Ecosystems.pptx
 
Energy is the beat of life irrespective of the domains. ATP- the energy curre...
Energy is the beat of life irrespective of the domains. ATP- the energy curre...Energy is the beat of life irrespective of the domains. ATP- the energy curre...
Energy is the beat of life irrespective of the domains. ATP- the energy curre...
 
POGONATUM : morphology, anatomy, reproduction etc.
POGONATUM : morphology, anatomy, reproduction etc.POGONATUM : morphology, anatomy, reproduction etc.
POGONATUM : morphology, anatomy, reproduction etc.
 
Cyathodium bryophyte: morphology, anatomy, reproduction etc.
Cyathodium bryophyte: morphology, anatomy, reproduction etc.Cyathodium bryophyte: morphology, anatomy, reproduction etc.
Cyathodium bryophyte: morphology, anatomy, reproduction etc.
 
COMPOSTING : types of compost, merits and demerits
COMPOSTING : types of compost, merits and demeritsCOMPOSTING : types of compost, merits and demerits
COMPOSTING : types of compost, merits and demerits
 
Porella : features, morphology, anatomy, reproduction etc.
Porella : features, morphology, anatomy, reproduction etc.Porella : features, morphology, anatomy, reproduction etc.
Porella : features, morphology, anatomy, reproduction etc.
 
GBSN - Biochemistry (Unit 3) Metabolism
GBSN - Biochemistry (Unit 3) MetabolismGBSN - Biochemistry (Unit 3) Metabolism
GBSN - Biochemistry (Unit 3) Metabolism
 
GBSN - Microbiology (Unit 5) Concept of isolation
GBSN - Microbiology (Unit 5) Concept of isolationGBSN - Microbiology (Unit 5) Concept of isolation
GBSN - Microbiology (Unit 5) Concept of isolation
 
Lipids: types, structure and important functions.
Lipids: types, structure and important functions.Lipids: types, structure and important functions.
Lipids: types, structure and important functions.
 
ONLINE VOTING SYSTEM SE Project for vote
ONLINE VOTING SYSTEM SE Project for voteONLINE VOTING SYSTEM SE Project for vote
ONLINE VOTING SYSTEM SE Project for vote
 
X-rays from a Central “Exhaust Vent” of the Galactic Center Chimney
X-rays from a Central “Exhaust Vent” of the Galactic Center ChimneyX-rays from a Central “Exhaust Vent” of the Galactic Center Chimney
X-rays from a Central “Exhaust Vent” of the Galactic Center Chimney
 
LUNULARIA -features, morphology, anatomy ,reproduction etc.
LUNULARIA -features, morphology, anatomy ,reproduction etc.LUNULARIA -features, morphology, anatomy ,reproduction etc.
LUNULARIA -features, morphology, anatomy ,reproduction etc.
 
Daily Lesson Log in Science 9 Fourth Quarter Physics
Daily Lesson Log in Science 9 Fourth Quarter PhysicsDaily Lesson Log in Science 9 Fourth Quarter Physics
Daily Lesson Log in Science 9 Fourth Quarter Physics
 
Concept of gene and Complementation test.pdf
Concept of gene and Complementation test.pdfConcept of gene and Complementation test.pdf
Concept of gene and Complementation test.pdf
 

Optimization of probabilistic argumentation with Markov processes

  • 1. optimization of probabilistic argumentation with markov processes E. Hadoux1 , A. Beynier1 , N. Maudet1 , P. Weng2 and A. Hunter3 Tue., Sept. 29th (1) Sorbonne Universités, UPMC Univ Paris 6, UMR 7606, LIP6, F-75005, Paris, France (2) SYSU-CMU Joint Institute of Engineering, Guangzhou, China SYSU-CMU Shunde International Joint Research Institute, Shunde, China (3) Department of Computer Science, University College London, Gower Street, London WC1E 6BT, UK
  • 2. Introduction ∙ Debate argumentation problems between two agents 1
  • 3. Introduction ∙ Debate argumentation problems between two agents ∙ Probabilistic executable logic to improve expressivity 1
  • 4. Introduction ∙ Debate argumentation problems between two agents ∙ Probabilistic executable logic to improve expressivity ∙ New class of problems: Argumentation Problem with Probabilistic Strategies (APS) (Hunter, 2014) 1
  • 5. Introduction ∙ Debate argumentation problems between two agents ∙ Probabilistic executable logic to improve expressivity ∙ New class of problems: Argumentation Problem with Probabilistic Strategies (APS) (Hunter, 2014) ∙ Purpose of this work: optimize the sequence of arguments of one agent 1
  • 6. Introduction ∙ Debate argumentation problems between two agents ∙ Probabilistic executable logic to improve expressivity ∙ New class of problems: Argumentation Problem with Probabilistic Strategies (APS) (Hunter, 2014) ∙ Purpose of this work: optimize the sequence of arguments of one agent There will be abuse of the word predicate! 1
  • 8. Formalization of a debate problem ∙ Turn-based game between two agents ∙ Rules to fire in order to attack arguments of the opponent and revise knowledge 3
  • 9. Formalization of a debate problem ∙ Turn-based game between two agents ∙ Rules to fire in order to attack arguments of the opponent and revise knowledge Let us define a debate problem with: ∙ A, the set or arguments 3
  • 10. Formalization of a debate problem ∙ Turn-based game between two agents ∙ Rules to fire in order to attack arguments of the opponent and revise knowledge Let us define a debate problem with: ∙ A, the set or arguments ∙ E, the set of attacks 3
  • 11. Formalization of a debate problem ∙ Turn-based game between two agents ∙ Rules to fire in order to attack arguments of the opponent and revise knowledge Let us define a debate problem with: ∙ A, the set or arguments ∙ E, the set of attacks ∙ P = 2A × 2E, the public space gathering voiced arguments 3
  • 12. Formalization of a debate problem ∙ Turn-based game between two agents ∙ Rules to fire in order to attack arguments of the opponent and revise knowledge Let us define a debate problem with: ∙ A, the set or arguments ∙ E, the set of attacks ∙ P = 2A × 2E, the public space gathering voiced arguments ∙ Two agents: agent 1 and agent 2 3
  • 13. Notation ∙ Arguments: literals (e.g., a, b, c) 4
  • 14. Notation ∙ Arguments: literals (e.g., a, b, c) ∙ Attacks: e(x, y) if x attacks y 4
  • 15. Notation ∙ Arguments: literals (e.g., a, b, c) ∙ Attacks: e(x, y) if x attacks y ∙ Args. in public (resp. private) space: a(x) (resp. hi(x)) 4
  • 16. Notation ∙ Arguments: literals (e.g., a, b, c) ∙ Attacks: e(x, y) if x attacks y ∙ Args. in public (resp. private) space: a(x) (resp. hi(x)) ∙ Goals: ∧ k g(xk) (resp. g(¬xk)) if xk is (resp. is not) accepted in the public space (Dung, 1995) 4
  • 17. Notation ∙ Arguments: literals (e.g., a, b, c) ∙ Attacks: e(x, y) if x attacks y ∙ Args. in public (resp. private) space: a(x) (resp. hi(x)) ∙ Goals: ∧ k g(xk) (resp. g(¬xk)) if xk is (resp. is not) accepted in the public space (Dung, 1995) ∙ Rules: prem ⇒ Pr(Acts) 4
  • 18. Notation ∙ Arguments: literals (e.g., a, b, c) ∙ Attacks: e(x, y) if x attacks y ∙ Args. in public (resp. private) space: a(x) (resp. hi(x)) ∙ Goals: ∧ k g(xk) (resp. g(¬xk)) if xk is (resp. is not) accepted in the public space (Dung, 1995) ∙ Rules: prem ⇒ Pr(Acts) ∙ Premises: conjunctions of e(, ), a(), hi() 4
  • 19. Notation ∙ Arguments: literals (e.g., a, b, c) ∙ Attacks: e(x, y) if x attacks y ∙ Args. in public (resp. private) space: a(x) (resp. hi(x)) ∙ Goals: ∧ k g(xk) (resp. g(¬xk)) if xk is (resp. is not) accepted in the public space (Dung, 1995) ∙ Rules: prem ⇒ Pr(Acts) ∙ Premises: conjunctions of e(, ), a(), hi() ∙ Acts: conjunctions of ⊞, ⊟ on e(, ), a() and ⊕, ⊖ on hi() 4
  • 20. Formalization of an APS An APS is characterized (from the point of view of agent 1) by ⟨A, E, G, S1, g1, g2, S2, P, R1, R2⟩: ∙ A, E, P as specified above ∙ G, the set of all possible goals ∙ Si, the set of private states for agent i ∙ gi ∈ G, the given goal for agent i ∙ Ri, the set of rules for agent i 5
  • 22. Example: Arguments Is e-sport a sport? a E-sport is a sport b E-sport requires focusing, precision and generates tiredness c Not all sports are physical d Sports not referenced by IOC exist e Chess is a sport f E-sport is not a physical activity g E-sport is not referenced by IOC h Working requires focusing and generates tiredness but is not a sport 6
  • 23. Example: Formalization ∙ A = {a, b, c, d, e, f, g, h} 7
  • 24. Example: Formalization ∙ A = {a, b, c, d, e, f, g, h} ∙ E = { e(f, a), e(g, a), e(b, f), e(c, f), e(h, b), e(g, c), e(d, g), e(e, g)} 7
  • 25. Example: Formalization ∙ A = {a, b, c, d, e, f, g, h} ∙ E = { e(f, a), e(g, a), e(b, f), e(c, f), e(h, b), e(g, c), e(d, g), e(e, g)} ∙ g1 = g(a) 7
  • 26. Example: Formalization ∙ A = {a, b, c, d, e, f, g, h} ∙ E = { e(f, a), e(g, a), e(b, f), e(c, f), e(h, b), e(g, c), e(d, g), e(e, g)} ∙ g1 = g(a) ∙ R1 = {h1(a) ⇒ ⊞a(a), h1(b) ∧ a(f) ∧ h1(c) ∧ e(b, f) ∧ e(c, f) ⇒ 0.5 : ⊞a(b) ∧ ⊞e(b, f) ∨ 0.5 : ⊞a(c) ∧ ⊞e(c, f), h1(d) ∧ a(g) ∧ h1(e) ∧ e(d, g) ∧ e(e, g) ⇒ 0.8 : ⊞a(e) ∧ ⊞e(e, g) ∨ 0.2 : ⊞a(d) ∧ ⊞e(d, g)} 7
  • 27. Example: Formalization ∙ R2 = {h2(h) ∧ a(b) ∧ e(h, b) ⇒ ⊞a(h) ∧ ⊞e(h, b), h2(g) ∧ a(c) ∧ e(g, c) ⇒ ⊞a(g) ∧ ⊞e(g, c), a(a) ∧ h2(f) ∧ h2(g) ∧ e(f, a) ⇒ 0.8 : ⊞a(f) ∧ ⊞e(f, a) ∨ 0.2 : ⊞a(g) ∧ ⊞e(g, a)} ∙ Initial state: h1(a, b, c, d, e), {}, h2(f, g, h) 8
  • 28. Attacks graph a g f c bde h Figure: Graph of arguments of Example e-sport 9
  • 29. Probabilistic Finite State Machine: Graph APS → Probabilistic Finite State Machine σ1start σ2 σ3 σ4 σ5 σ6 σ7 σ8 σ9 σ10 σ11 σ12 1 0.8 0.2 0.5 0.5 1 1 0.8 0.2 0.8 0.2 Figure: PFSM of Example e-sport 10
  • 30. Probabilistic Finite State Machine To optimize the sequence of arguments for agent 1, we could optimize the PFSM but: 11
  • 31. Probabilistic Finite State Machine To optimize the sequence of arguments for agent 1, we could optimize the PFSM but: 1. depends of the initial state 11
  • 32. Probabilistic Finite State Machine To optimize the sequence of arguments for agent 1, we could optimize the PFSM but: 1. depends of the initial state 2. requires knowledge of the private state of the opponent 11
  • 33. Probabilistic Finite State Machine To optimize the sequence of arguments for agent 1, we could optimize the PFSM but: 1. depends of the initial state 2. requires knowledge of the private state of the opponent Using Markov models, we can relax assumptions 1 and 2. Moreover, the APS formalization can be modified in order to comply with the Markov assumption. 11
  • 34. Markov Decision Process A Markov Decision Process (MDP) (Puterman, 1994) is characterized by a tuple ⟨S, A, T, R⟩: ∙ S, a set of states, ∙ A, a set of actions, ∙ T : S × A → Pr(S), a transition function, ∙ R : S × A → R, a reward function. 12
  • 35. Partially-Observable Markov Decision Process A Partially-Observable MDP (POMDP) (Puterman, 1994) is characterized by a tuple ⟨S, A, T, R, O, Q⟩: ∙ S, a set of states, ∙ A, a set of actions, ∙ T : S × A → Pr(S), a transition function, ∙ R : S × A → R, a reward function, ∙ O, an observation set, ∙ Q : S × A → Pr(O), an observation function. 13
  • 36. Mixed-Observability Markov Decision Process A Mixed-Observability MDP (MOMDP) (Ong et al., 2010) is characterized by a tuple ⟨Sv, Sh, A, T, R, Ov, Oh, Q⟩: ∙ Sv, Sh, a visible and hidden parts of the state, ∙ A, a set of actions, ∙ T : Sv × A × Sh → Pr(Sv × Sh), a transition function, ∙ R : Sv × A × Sh → R, a reward function, ∙ Ov = Sv, an observation set on the visible part of the state, ∙ Oh, an observation set on the hidden part of the state, ∙ Q : Sv × A × Sh → Pr(Ov × Oh), an observation function. 14
  • 38. Transformation to a MOMDP An APS from the point of view of agent 1 can be transformed to a MOMDP: ∙ Sv = S1 × P, Sh = S2 ∙ A = {prem(r) ⇒ m|r ∈ R1 and m ∈ acts(r)} ∙ Ov = Sv and Oh = ∅ ∙ Q(⟨sv, sh⟩, a, ⟨sv⟩) = 1, otherwise 0 ∙ T, see after 16
  • 39. Transformation to a MOMDP: Transition function Application set Let Cs(Ri) be the set of rules of Ri that can be fired in state s. The application set Fr(m, s) is the set of predicates resulting from the application of act m of a rule r on s. If r cannot be fired in s, Fr(m, s) = s. ∙ s, a state and r : p ⇒ m, an action s.t. r ∈ A ∙ s′ = Fr(m, s) ∙ r′ ∈ Cs′ (R2) s.t. r′ : p′ ⇒ [π1/m1, . . . , πn/mn] ∙ s′′ i = Fr′ (mi, s′) ∙ T(s, r, s′′ i ) = πi 17
  • 40. Reward function For the reward function: ∙ with Dung’s semantics: positive reward for each part holding ∙ can be generalized: General Gradual Valuation (Cayrol and Lagasquie-Schiex, 2005) 18
  • 41. Transformation to a MOMDP Model sizes: APS : 8 arguments, 8 attacks, 6 rules POMDP : 4 294 967 296 states MOMDP : 16 777 216 states Untractable instances → need to optimize at the root 19
  • 43. Solving an APS Two algorithms to solve MOMDPs: ∙ MO-IP (Araya-López et al., 2010), IP of POMDP on MOMDP (exact method) ∙ MO-SARSOP (Ong et al., 2010), SARSOP of POMDP on MOMDP (approximate method albeit very efficient) Two kinds of optimizations: with or without dependencies on the initial state 21
  • 44. Optimizations without dependencies Irr. Prunes irrelevant arguments 22
  • 45. Optimizations without dependencies Irr. Prunes irrelevant arguments Enth. Infers attacks 22
  • 46. Optimizations without dependencies Irr. Prunes irrelevant arguments Enth. Infers attacks Dom. Removes dominated arguments 22
  • 47. Optimizations without dependencies Irr. Prunes irrelevant arguments Enth. Infers attacks Dom. Removes dominated arguments Guarantee on the unicity and optimality of the solution. 22
  • 48. Attacks graph Argument dominance If an argument is attacked by any unattacked argument, it is dominated. a f g b c d e h Figure: Attacks graph of Example 23
  • 49. Optimization with dependencies Irr(s0) has to be reapplied each time the initial state changes. 24
  • 50. Optimization with dependencies Irr(s0) has to be reapplied each time the initial state changes. 1. For each predicate that is never modified but used as premises: 1.1 Remove all the rules that are not compatible with the value of this predicate in the initial state. 1.2 For all remaining rules, remove the predicate from the premises. 24
  • 51. Optimization with dependencies Irr(s0) has to be reapplied each time the initial state changes. 1. For each predicate that is never modified but used as premises: 1.1 Remove all the rules that are not compatible with the value of this predicate in the initial state. 1.2 For all remaining rules, remove the predicate from the premises. 2. For each remaining action of agent 1, track the rules of agent 2 compatible with the application of this action. If a rule of agent 2 is not compatible with any application of an action of agent 1, remove it. 24
  • 53. Experiments We computed a solution for the e-sport problem with: ∙ MO-IP, which did not finish after tens of hours ∙ MO-SARSOP without optimizations, idem ∙ MO-SARSOP with optimizations, 4sec for the optimal solution 26
  • 54. Experiments: Policy graph r1 1,1start r1 2,2 r1 3,1 ∅ r1 3,1∅ r1 2,2 ∅ ∅ o2 o5o4 o6 o7 o8 o5 o1 o7 o8 o3 o3 o4 Figure: Policy graph for Example 27
  • 55. Experiments: More examples None Irr. Enth. Dom. Irr(s0). All Ex 1 — — — — — 0.56 Ex 2 3.3 0.3 0.3 0.4 0 0 Dv. — — — — — 32 6 1313 22 43 7 2.4 0.9 7 — 180 392 16 20 6.7 8 — — — — 319 45 9 — — — — — — Table: Computation time (in seconds) 28
  • 57. Conclusion We presented: 1. A new framework to represent more complex debate problems (APS) 2. A method to transform those problems to a MOMDP 3. Several optimizations that can be used outside of the context of MOMDP 4. A method to optimize actions of an agent in an APS 30
  • 58. Perspectives We are currently working on using POMCP (Silver and Veness, 2010). We are also using HS3MDPs (Hadoux et al., 2014). 31
  • 60. Bibliography I Araya-López, M., Thomas, V., Buffet, O., and Charpillet, F. (2010). A closer look at MOMDPs. In 22nd IEEE International Conference on Tools with Artificial Intelligence (ICTAI). Cayrol, C. and Lagasquie-Schiex, M.-C. (2005). Graduality in argumentation. Journal of Artificial Intelligence Research (JAIR), 23:245–297. Dung, P. M. (1995). On the acceptability of arguments and its fundamental role in nonmonotonic reasoning, logic programming and n-person games. Artificial Intelligence, 77(2):321–358. 33
  • 61. Bibliography II Hadoux, E., Beynier, A., and Weng, P. (2014). Solving Hidden-Semi-Markov-Mode Markov Decision Problems. In Straccia, U. and Calì, A., editors, Scalable Uncertainty Management, volume 8720 of Lecture Notes in Computer Science, pages 176–189. Springer International Publishing. Hunter, A. (2014). Probabilistic strategies in dialogical argumentation. In International Conference on Scalable Uncertainty Management (SUM’14) LNCS volume 8720. Ong, S. C., Png, S. W., Hsu, D., and Lee, W. S. (2010). Planning under uncertainty for robotic tasks with mixed observability. In The International Journal of Robotics Research. 34
  • 62. Bibliography III Puterman, M. L. (1994). Markov Decision Processes: discrete stochastic dynamic programming. John Wiley & Sons. Silver, D. and Veness, J. (2010). Monte-Carlo planning in large POMDPs. In Proceedings of the 24th Conference on Neural Information Processing Systems (NIPS), pages 2164–2172. 35