Companions Florian PECUNE Magalie OCHS Catherine PELACHAUD CNRS LTCI Télécom Paristech The MoCA Project Florian Pecune page 1 Create a little world of artificial companions ID: 1045802
Download Presentation The PPT/PDF document "A Formal Model of Social Relations for A..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
1. A Formal Model of Social Relations for Artificial CompanionsFlorian PECUNE – Magalie OCHS – Catherine PELACHAUDCNRS – LTCI, Télécom Paristech
2. The MoCA ProjectFlorian Pecunepage 1Create a “little world” of artificial companions…… with different roles and behavior“…a robot or a virtual conversational agent that possesses […] social skills that allow it to establish and maintain long-term relationships with users.” (Lim, 2012)Focus on dyadic interactions
3. Thesis ObjectiveFlorian Pecunepage 2Create an agent able to make decisions according to its relation … … but also to influence user’s relationHow to represent social relations?Three research problematics:How do they change during interactions?I do that because I like the userI do that because I want the user to like meHow to model the effects of these relations?
4. Florian Pecunepage 3Ideal Social RelationExpressed Social RelationRelation the agent would like to expressIntrinsic relation felt by the agentRelation expressed by the agentFelt Social RelationI like the user, I am less dominant than him I would like to be more dominant than the user Influenced by agent’s personality (Trapnell & Wiggins, 1990)Influenced by agent’s role (Heise, 1987)Influenced by felt and ideal relations (Scherer, 2005)Influenced by Mood (Isen & al., 1992)How to represent social relations? (1/2)
5. Florian Pecunepage 4Ideal Social RelationExpressed Social RelationRelation the agent would like to expressIntrinsic relation felt by the agentRelation expressed by the agentFelt Social RelationI like the user, I am less dominant than him I would like to be more dominant than the user Influenced by agent’s personality (Trapnell & Wiggins, 1990)Influenced by agent’s role (Heise, 1987)Influenced by felt and ideal relations (Scherer, 2005)Influenced by Mood (Isen & al., 1992)How to represent social relations? (1/2)
6. How to represent social relations? (2/2)Florian Pecunepage 5DominanceFamiliarity“the capacity of one agent to affect the behavior of another” (Prada & Paiva, 2008)“mutual knowledge of personal information” (Svennevig, 1999)DominantLikingDislikingSubmissiveSeveral dimensions used to define social relations Dominance and liking are the most widely used (Argyle, 1988)Familiarity useful for long-term relationship (Bickmore & Picard, 2005)(Altman & Taylor, 1973)Liking“A general and enduring positive or negative feeling about some person” (Moshkina & Arkin, 2003)
7. Dominance – The dependence (1/3)Florian Pecunepage 6Dependence of an agent A upon an agent B (Emerson, 1962) “Directly proportional to A's motivational investment in goals mediated by B (…)(…) Inversely proportional to the availability of those goals to A outside of the A-B relation”Agent BAgent AI really want to go to the concert tonight !High DependenceAgent BAgent AI’m the only one who can bring you there.Agent BAgent AMaybe I could go to the concert tonight.Low DependenceAgent BAgent AI really want to go to the concert tonight !I can bring you there with my car.We can bring you there too !I can bring you there with my car.I can bring you there with my car.I really want to go to the concert tonight !
8. Dominance – The dependence (2/3)Florian Pecunepage 7An agent i can be helpful ….… or threateningAgent BAgent AI really want to go to the concert tonight !I can bring you there with my car.Agent BAgent AI really want to go to the concert tonight !Not if I cancel the concert !If j has an intention And i can do an action a after what it will be impossible to achieve And i can do an action a after what will be true If j has an intention Or i can do an action a’ cancelling a precondition of
9. Dominance – The dependence (3/3)Florian Pecunepage 8j is dependant upon i toward an intention i can help j to achieve this intentionj can not achieve this intention himselfi can prevent j to achieve this intention andormeans that j believes
10. Dominance – The dependence value (1/2)Florian Pecunepage 9Importance accorded by i to the success of an intention Dependence influenced by the importance accorded to intentions… Importance accorded by i to the non failure of an intention … and the number of potential helpersNumber of agents susceptible to have a positive impact on the intention The number of potential threatening agents is not taken into account Agent AIt is normal to be understood by the userLow importance of successAgent AIt is not normal if the user do not understand me !High importance of failure
11. Dominance – The dependence value (2/2)Florian Pecunepage 10Dependence valuePositively influenced by the importance accorded to the intention (success or non failure)Negatively influenced by the number of potential helpersAgent BAgent AI really want to go to the concert tonight !I can bring you there with my car.High DependenceHigh importanceLow numberAgent BAgent AI don’t care that muchI can destroy your TV !Low importanceLow Dependence
12. Dominance – Dominance valueFlorian Pecunepage 11i dominance toward jBi(j dependence upon i)Bi(i dependence upon j)The dominance value of an agent i toward another agent j is :The difference between his belief about j’s dependence upon him and his believed dependence upon jAgent BAgent AI really want to go to the concert tonight !I can bring you there with my car.High DependenceAgent BAgent AI can destroy your TV!I don’t care that much…Low DependenceB is dominantThe agent i is dominant toward j if he believes he is less dependant upon j than j is upon him
13. Dominance – Strategies (1/2)Florian Pecunepage 12To improve his level of dominance, an agent i can : Decrease the importance accorded to one of his intentions that the agent j may achieve or threatenDecrease his dependency by finding other sources able to help himAgent BAgent AI really want to go to the concert tonight !I can bring you there with my car.High importanceLow numberAgent BAgent AFinally, I don’t care…I can bring you there with my car.Low numberLow importanceHigh numberAgent BAgent AI really want to go to the concert tonight !I can bring you there with my car.We can bring you there too !High importance
14. Dominance – Strategies (2/2)Florian Pecunepage 13To improve his level of dominance, an agent i can : Try to increase the motivational investment of the other agentDeny alternative sources for the other agenti dominance toward jBi(j dependence upon i)Bi(i dependence upon j)Find other intentions for which he can be helpfulFind other intentions for which he can be threatening
15. How to represent social relations? (2/2)Florian Pecunepage 14DominanceFamiliarity“the capacity of one agent to affect the behavior of another” (Prada & Paiva, 2008)“mutual knowledge of personal information” (Svennevig, 1999)DominantLikingDislikingSubmissiveSeveral dimensions used to define social relations Dominance and liking are the most widely used (Argyle, 1988)Familiarity useful for long-term relationship (Bickmore & Picard, 2005)(Altman & Taylor, 1973)Liking“A general and enduring positive or negative feeling about some person” (Moshkina & Arkin, 2003)
16. Liking – TheoryFlorian Pecunepage 15Based on Heider’s Balance Theory(Heider, 1958) Concept XAgent BAgent AA relation is balanced : “If all three relations are positive in all respects (…)(….) or if two are negative and one positive.”ConceptAgent BAgent AConceptAgent BAgent AI like chocolate.Me too !I don’t like bravery.I do.People tend to seek balanced states (Zajonc, 1960)Sharing appreciations will improve the liking
17. Liking – AppreciationFlorian Pecunepage 16The agent i appreciate the concept XShared appreciationConceptAgent BAgent AConceptAgent BAgent AI like chocolate.Me too !I don’t like bravery.I do.Contradictory appreciation
18. Liking – Liking valueFlorian Pecunepage 17i liking toward jImportance accorded to shared conceptsImportance accorded to contradictory conceptsImportance accorded to a concept XConceptAgent BAgent AI like chocolate.Me too !ConceptAgent BAgent AI don’t like bravery.I do.Low likingHigh dislikingLiking value of an agent i toward an agent j :
19. Liking – StrategiesFlorian Pecunepage 18To improve his level of liking, an agent i can : Change its own appreciation about a concept for which j has a different feelingTry to change j’s feeling about this same conceptConceptAgent BAgent AI don’t like football.I do.UnbalancedConceptAgent BAgent AI like football.Me too.ConceptAgent BAgent AI don’t like football.Me neither.
20. How to represent social relations? (2/2)Florian Pecunepage 19DominanceFamiliarity“the capacity of one agent to affect the behavior of another” (Prada & Paiva, 2008)“mutual knowledge of personal information” (Svennevig, 1999)DominantLikingDislikingSubmissiveSeveral dimensions used to define social relations Dominance and liking are the most widely used (Argyle, 1988)Familiarity useful for long-term relationship (Bickmore & Picard, 2005)(Altman & Taylor, 1973)Liking“A general and enduring positive or negative feeling about some person” (Moshkina & Arkin, 2003)
21. Familiarity – Theory Florian Pecunepage 20FamilyWorkLovePoliticsReligionSuperficialIntimatePersonalCoreBased on Social Penetration Theory(Altman & Taylor, 1973) Breadth represents the number of topics shared by the agentsDepth represents the intimacy value accorded to a topic
22. Familiarity – Theory Florian Pecunepage 21Breadth Ratio between the number of i’s beliefs about j’s feelings and the number of possible conversational topicsFamilyWorkLovePoliticsReligionSuperficialIntimatePersonalCoreAgent AHigh BreadthExcept for love, I talked with j about everything I would like to know
23. Familiarity – Theory Florian Pecunepage 22Depth FamilyWorkLovePoliticsReligionSuperficialIntimatePersonalCoreIntimacy value accorded to a concept X Low DepthMean value of Intimacy
24. Familiarity – Strategies Florian Pecunepage 23To improve his level of familiarity, an agent i can : Improve breadth by asking information about topics for which he does not have any beliefsImprove depth by asking more intimate information about a particular topic Agent ALet’s talk about Love !Agent ATell me more about your family.
25. Future WorkFlorian Pecunepage 24Define scenarios to test this modelFormalize influence of personality on social relationsImplement a neural network to represent the dynamics of the relationsFormalize ideal and expressed social relations
26. A Formal Model of Social Relations for Artificial CompanionsFlorian PECUNE – Magalie OCHS – Catherine PELACHAUDCNRS – LTCI, Télécom Paristech
27. Logical FrameworkFlorian Pecunepage 26Mental attitudes Operators The agent i believes that is true The agent i has the intention and actively desires to achieve it The action a might happen and will be true after thatcan not be true if is not already trueThe agent i can be the author of the action a