A Production Rule-Based Framework  for Causal and Epistemic Reasoning       Theodore Patkos, Abdelghani Chibani,       Dim...
Outline     • Overview     • (Epistemic) Action Theories     • Production Rule-based Framework     • Application Domain   ...
Motivation and Objectives    • Action theories and production systems have widely been used in KR&R to      represent know...
Framework overview                     Event Calculus                        DECKT  foundations  Theoretical              ...
Outline     • Overview     • (Epistemic) Action Theories     • Production Rule-based Framework     • Application Domain   ...
Background – Action Theories    • The objective is to express the dynamics of the world    • Therefore always include a mo...
Background –     Commonsense phenomena    • Related issues       • Representation       • Effects of Events       • Indire...
Background –       Discrete Time Event Calculus     • The EC distinguishes three kinds of objects –       events, fluents ...
Epistemic Action Theories     • Epistemic (modal) logic: An agent is       said to know a fact if this is true in       al...
Epistemic and Causal Reasoning     • Action theories that do not model time explicitly have been extended to       reason ...
DECKT –         Hidden Causal Dependencies      • We developed a unified formal theory for epistemic, causal and temporal ...
Outline     • Overview     • (Epistemic) Action Theories     • Production Rule-based Framework     • Application Domain   ...
EC Jess-based ReasonerRuleML 12                       13
Requirements and Challenges    • There exist different implementations of the Event Calculus for offline      reasoning,  ...
DECKT Time-Dependent Meta-Axioms    • Event e initiates f if f’ is true    • (KT6.1.1) handles the case of      unknown pr...
Production rules with high level     structures    • Features that we employ with rule-based reasoning involve       • Dyn...
Operational semantics –     Model generator                                  KB                    Sensing and        DEC/...
Non-Epistemic Reasoning Cycle              T=t                  T=t+1                            i   [HoldsAt(fi,t)]      ...
Epistemic Reasoning Cycle            T=t                T=t+1     KB of   Epistemic    Fluents                            ...
An Informal Note on Complexity    • The logical theories set a lower bound as to how efficient an      implementation can ...
Framework overview                     Event Calculus                        DECKT  foundations  Theoretical              ...
Outline     • Overview     • (Epistemic) Action Theories     • Production Rule-based Framework     • Application Domain   ...
Application Domain – Ambient     Intelligence and Ubiquitous Robotics     • Sensor-rich collaborative environments     • T...
Conceptual Layers – Knowledge      representation in a smart space                           • Moving from low-level      ...
High-level Activity Recognition –     Challenges for the State-of-the-Art    • Semantic Web tools (ontologies and rule-bas...
Reactive Reasoning and      Temporal ProjectionRuleML 12                      26
Outline     • Overview     • (Epistemic) Action Theories     • Production Rule-based Framework     • Application Domain   ...
Contributions and     Ongoing Work    • The Event Calculus provides a declarative specification of state transitions,     ...
The end     Thank you for your attention!
Upcoming SlideShare
Loading in …5
×

Ruleml2012 - A production rule-based framework for causal and epistemic reasoning

848 views

Published on

Published in: Technology, Education
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
848
On SlideShare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
11
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Ruleml2012 - A production rule-based framework for causal and epistemic reasoning

  1. 1. A Production Rule-Based Framework for Causal and Epistemic Reasoning Theodore Patkos, Abdelghani Chibani, Dimitris Plexousakis, Yacine Amirat ({patkos, chibani, amirat}@u-pec.fr, dp@ics.forth.gr) 6th International Symposium on Rules (RuleML’12), Laboratoire Images Signaux et Systèmes Intelligents (LISSI) University of Paris-Est Creteil, Paris Institute of Computer Science – Foundation for Research and Technology Hellas (FO.R.T.H.)Research Work supported by the ITEA 2 EU Projects: A2NETS, PREDYKOT
  2. 2. Outline • Overview • (Epistemic) Action Theories • Production Rule-based Framework • Application Domain • ConclusionsRuleML 12 2
  3. 3. Motivation and Objectives • Action theories and production systems have widely been used in KR&R to represent knowledge change in dynamic domains. • Our objective is to exploit the expressive capacity of logic-based theories and the efficiency of rule-based systems, reconciling their differences. • The outcome is a complete framework that performs runtime reasoning about events, knowledge and time in expressive domains. • The system can carry out temporal projection and deductive narrative verification tasks and is evaluated in real-world settingsRuleML 12 3
  4. 4. Framework overview Event Calculus DECKT foundations Theoretical • Reasoning about action and time • Epistemic reasoning • Solution to problems (frame, • Hidden causal dependencies, rather ramification, qualification) than possible worlds structures • Commonsense phenomena • Sensing, potential actions etc Implementation Rule-based forward-chaining production system • NaF, semi-destructive update • Salience values, subsumption… • Application Domain Contribution Online/offline reasoning • Multiple model • Ambient Intelligence, AAL generation • Benchmark problems (e.g., • GUI/Java interface Shanahan’s circuit)RuleML 12 4
  5. 5. Outline • Overview • (Epistemic) Action Theories • Production Rule-based Framework • Application Domain • ConclusionsRuleML 12 5
  6. 6. Background – Action Theories • The objective is to express the dynamics of the world • Therefore always include a more or less implicit general notions of time, change and causality. • Action Theories automate the process of commonsense reasoning, in order to • predict the outcome of a given action sequence • explain observations • find a situation in which certain goal conditions are met.RuleML 12 6
  7. 7. Background – Commonsense phenomena • Related issues • Representation • Effects of Events • Indirect Effects of Events (Ramification problem) • Context-dependent Effects • Non-deterministic Effects • Concurrent Events • Preconditions • Inertia (Frame problem) • Actions with duration • Delayed Effects and Continuous Change • Default Reasoning (Qualification problem) • …RuleML 12 7
  8. 8. Background – Discrete Time Event Calculus • The EC distinguishes three kinds of objects – events, fluents and timepoints. To do this, it appeals to a sorted first-order language. EC Predicates • Commonsense Law of Inertia: things tend to • HoldsAt(f,t) persist unless affected by some event. • ReleasedAt(f,t) • Happens(e,t) Positive and Negative Effect Axioms (Σ) fi C [HoldsAt(f ,t)] Initiates(e,f,t) i fi C [HoldsAt(f ,t)] Terminates(e,f,t) • Initiates(e,f,t) i • Terminates(e,f,t) State Constraints (Ψ) • Releases(e,f,t) i [HoldsAt(f ,t)] HoldsAt(f,t) iRuleML 12 8
  9. 9. Epistemic Action Theories • Epistemic (modal) logic: An agent is said to know a fact if this is true in all possible worlds.RuleML 12 9
  10. 10. Epistemic and Causal Reasoning • Action theories that do not model time explicitly have been extended to reason about knowledge in a straightforward manner. • The Situation Calculus [Moore 1985, Scherl&Levesque 2003] • The Fluent Calculus [Thielscher 2000] • The Action Language Ak [Lobo et al. 2001] • For example, suppose an action E that makes F true if F’ is true:RuleML 12 10
  11. 11. DECKT – Hidden Causal Dependencies • We developed a unified formal theory for epistemic, causal and temporal reasoning • able to express diverse phenomena of E commonsense reasoning, about knowledge • still being computationally feasible.RuleML 12Theodore Patkos et al. 11
  12. 12. Outline • Overview • (Epistemic) Action Theories • Production Rule-based Framework • Application Domain • ConclusionsRuleML 12 12
  13. 13. EC Jess-based ReasonerRuleML 12 13
  14. 14. Requirements and Challenges • There exist different implementations of the Event Calculus for offline reasoning, • but certain of its features are not appropriate for runtime execution (e.g., explicit frame axioms – computational frame problem) • DECKT has been designed for practical implementations, • Therefore introduces added features not typically met in Event Calculus (e.g., reification of formulas, time-dependent meta axioms for handling of HCDs)RuleML 12 14
  15. 15. DECKT Time-Dependent Meta-Axioms • Event e initiates f if f’ is true • (KT6.1.1) handles the case of unknown preconditions e • Circumscription at every timepoint would be inefficient. fi C [HoldsAt(f ,t)] Initiates(e,f,t) iRuleML 12 15
  16. 16. Production rules with high level structures • Features that we employ with rule-based reasoning involve • Dynamic rule construction to accommodate HCDs • Semi-destructive update of the KB, where applicable • NaF, rather than circumscription • List handling • Salience values for conflict resolution (e.g., concurrent events). • Numerical manipulationsRuleML 12 16
  17. 17. Operational semantics – Model generator KB Sensing and DEC/DECKT axioms Acting (destructive update) Ramifications Alternative (state constraints) Models Triggered events • Released fluents and NaFRuleML 12 17
  18. 18. Non-Epistemic Reasoning Cycle T=t T=t+1 i [HoldsAt(fi,t)] HoldsAt(f,t) KB of Fluents Released . Trigger . . X State axioms Sensing Constrains Released . . . X Effect Constrained axioms . . . Released fi C[HoldsAt(f ,t)] i fi C [HoldsAt(f ,t)] i Initiates(e,f,t) Terminates(e,f,t) fi C [HoldsAt(f1,t)] Happens(e,t) XRuleML 12 18
  19. 19. Epistemic Reasoning Cycle T=t T=t+1 KB of Epistemic Fluents Trigger State axioms Effect Constrains Sensing axiomsRuleML 12 19
  20. 20. An Informal Note on Complexity • The logical theories set a lower bound as to how efficient an implementation can be. • The predominant computational complexity factor… • for the non-epistemic case is the number of released fluents • for the epistemic case is the set of HCDs • Query answering on ground facts is of linear complexity. • Jess pattern matching depends on the syntactic form of rules. • Ranges from O(p) to O(pn)RuleML 12 20
  21. 21. Framework overview Event Calculus DECKT foundations Theoretical • Reasoning about action and time • Epistemic reasoning • Solution to problems (frame, • Hidden causal dependencies, rather ramification, qualification) than possible worlds structures • Commonsense phenomena • Sensing, potential actions etc Implementation Rule-based forward-chaining production system • NaF, semi-destructive update • Salience values, subsumption… • Application Domain Contribution Online/offline reasoning • Multiple model • Ambient Intelligence, AAL generation • Benchmark problems (e.g., • GUI/Java interface Shanahan’s circuit)RuleML 12 21
  22. 22. Outline • Overview • (Epistemic) Action Theories • Production Rule-based Framework • Application Domain • ConclusionsRuleML 12 22
  23. 23. Application Domain – Ambient Intelligence and Ubiquitous Robotics • Sensor-rich collaborative environments • Temporal constraints are ubiquitousRuleML 12 23
  24. 24. Conceptual Layers – Knowledge representation in a smart space • Moving from low-level data to high-level knowledge inference more expressive tools are needed • AI has a decisive role to play: • representation of contextual knowledge, • context inference, • collaboration of devices to achieve common objectives, • planning in dynamic domains, • commonsense reasoningRuleML 12 24
  25. 25. High-level Activity Recognition – Challenges for the State-of-the-Art • Semantic Web tools (ontologies and rule-based reasoning) are widely used to tackle AmI-related problems • Complex ambient systems test the limits of these methods in terms of expressiveness and reasoning capacity. • Agents inhabiting smart spaces need to exhibit • Cognitive skills and commonsense reasoning • Temporal reasoning • Operate under partial observabilityRuleML 12 25
  26. 26. Reactive Reasoning and Temporal ProjectionRuleML 12 26
  27. 27. Outline • Overview • (Epistemic) Action Theories • Production Rule-based Framework • Application Domain • ConclusionsRuleML 12 27
  28. 28. Contributions and Ongoing Work • The Event Calculus provides a declarative specification of state transitions, while DECKT provides a Kripke-equivalent epistemic semantics • Production rules obtain high-level structures. • We aim at a tool for both educational and practical use. • We extend the editor to support the full expressive power of the EC • Benchmark and use case evaluation of real settings is our ongoing work, considering further enhancements of the system • We study its integration with probabilistic methods of inference.RuleML 12 28
  29. 29. The end Thank you for your attention!

×