Y. Wilks and A. Ballim. Multiple agents and the heuristic ascription of belief. In Proc. of the 10th International Joint Conference on Artificial Intelligence, pages 118--124, 1987.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....for the representation of propositional attitudes in the 2 implementation of arti cial reasoners. In fact, ML systems better resemble the structure of some of the most successful existing computer systems. One example in the representation of propositional attitudes is Wilks system ViewGen [47, 49] which allows the use of explicitly distinct sets of beliefs (each set with its own signature) Analogously, in the area of theorem proving, the OYSTER CLAM system [28, 43] which uses an explicit declarative metalevel [3] has a metatheory which is distinct from the object theory and has also a ....

Y. Wilks and A. Ballim. Multiple agents and the heuristic ascription of belief. In Proc. of the 10th International Joint Conference on Articial Intelligence, pages 118-124, 1987.

....be why his implementation of a solution for the Wise Man puzzle is very specifically geared towards the puzzle and does not generalize to other simulative reasoning problems. 3. 7 ViewGen ViewGen is a program that implements the theory of belief ascription developed by Ballim and Wilks et al. [Wilks and Ballim, 1987; Ballim et al. 1991; Ballim and Wilks, 1991; Ballim, 1992] 2 Note that hPS hP; S i; red; estari 6= red(estar) because the terms are of different type (n tuple vs. truth value) even though the proposition surrogate evaluates to red(estar) by the rules of lambda conversion. 33 ....

Y. Wilks and A. Ballim. Multiple agents and the heuristic ascription of belief. In Proceedings of the Tenth International Conference on Artificial Intelligence, pages 118--124, Palo Alto, CA, 1987. Morgan Kaufmann.

....in parallel with very little cross fertilization. Let us further consider the mechanization of propositional attitudes. In practical applications, the most common approach is to represent the beliefs of an agent as a set of facts plus some way to deduce consequences from them (see for instance [43, 44, 45]) Until recently, this and the logic based work (e.g. modal logics or McCarthy s formalism) were developed independently, the first lacking theoretical foundations, the second being hardly applicable in real systems. This problem is fixed in [46] which, via an equivalence result, shows that the ....

Y. Wilks and A. Ballim. Multiple agents and the heuristic ascription of belief. In Proc. of the 10th International Joint Conference on Artificial Intelligence, pages 118--124, 1987.

....standard practice in software specification, we 1 This notion will be made precise below. 2 As a matter of fact, our proposed framework formalizes ideas which have been exploited in many complex applications developed in various areas of Artificial Intelligence, e.g. computational linguistics [7,27,28], the formalization of opacity and transparency in belief contexts [1] the integration of information coming from heterogeneous data bases [24] planning[20] and multiagent systems [17,19,26,4,5] For the sake of completeness, it must also be said that our approach shares some intuitions with ....

Y. Wilks and A. Ballim. Multiple agents and the heuristic ascription of belief. In Proc. of the 10th International Joint Conference on Artificial Intelligence, pages 118--124, 1987.

....discrepancy and intention ascription as theory formation. 6.4. 3 Handling Model Discrepancies The philosophy behind ascribing default models to other world entities is similar to that which lies behind the use of the default ascriptional rule in the ViewGen system of Wilks and Ballim [WB87, page 119] namely assume that one s view of another entity s view is the same as one s own except [when] there is explicit evidence to the contrary. To an agent which is modelling its environment, the evidence that some entity is not behaving as expected will manifest itself in the form of a ....

....point of view, obey regulations has a lower priority than its main achievement goal reach destination. More sophisticated reasoning about agent behaviour would certainly benefit from a more sophisticated treatment of agent beliefs in particular, representation [Kon83] default ascription [WB87] revision [DW90] and agent desires for example, values and likes [KR92] hedonic states [Gre87, page 308] and goal commitment [CL87] As modelling capabilities are extended, however, agents will be faced with many more choices when it comes to explaining other agents behaviour. Since ....

Yorick Wilks and Afzal Ballim. Multiple agents and the heuristic ascription of belief. In Proceedings International Joint Conference on Artificial Intelligence, pages 118--124, 1987.

....for the representation of propositional attitudes in the implementation of artificial reasoners. In fact, ML systems better resemble the structure of some of the most successful existing computer systems. One example in the representation of propositional attitudes is Wilks system ViewGen [47, 49] which allows the use of explicitly distinct sets of beliefs (each set with its own signature) Analogously, in the area of theorem proving, the OYSTER CLAM system [28, 43] which uses an explicit declarative metalevel [3] has a metatheory which is distinct from the object theory and has also a ....

Y. Wilks and A. Ballim. Multiple agents and the heuristic ascription of belief. In Proc. of the 10th International Joint Conference on Artificial Intelligence, pages 118--124, 1987.

....Some subset of B c might be distinguished as common knowledge and hence assigned to B cs . We might also attempt to make some apriori ascriptions of beliefs to the student. For example, the system might assume that she knows nothing of what she is to learn: p ( p e O c B c B s p ) Wilks and Ballim (1987) suggest a general default belief ascription axiom: B a p q (B a q q B a B b p) B a B b p i.e. if an agent believes a proposition p and believes there is no reason q why a second agent should not believe it then assume that it does. However, they also point out that there are ....

.... we find an increasing emphasis on the student s ability to negotiate both about the concepts to be discussed and about the meaning of the concepts themselves (Moyse and Elsom Cook, 1991) To illustrate the potential relevance of viewpoints to student modelling we may adapt an example from Wilks and Ballim (1987), who consider a viewpoint to be an agent s set of 33 beliefs about some topic. The example emphasises that such viewpoints need to be nested. Imagine the system is mediating an interaction between two medical students a and b. We might have: B c = Type(thalassemia,genetic disorder) ....

Wilks, Y. and Ballim, A. (1987). Multiple agents and the heuristic ascription of beliefs, Proceedings of the 10th International Joint Conference on Artificial Intelligence, 118124.

....and Kelleher, 1988) Cooperative dialogues Since, we have argued, the construction of new beliefs occurs through criticism, the role of dialogue is crucial in ITSs. The main technical requirement is a means of handling nested beliefs (that is, beliefs about another agent s beliefs) For example, Wilks and Ballim (1987) present a formalisation of an interaction which may be re interpreted as one between a tutor (a doctor) and two students, one of whom believes that thalassemia is a genetic disorder , and the other of whom does not. The derivation of an appropriate tutor response to a remark from the first ....

Wilks, Y. and Ballim, A. (1987). Multiple agents and the heuristic ascription of belief. Proceedings of the International Joint Conference on Artificial Intelligence, Milan.

.... agent believes all the formulae which explicitly are in the context of its own beliefs) The framework of ML systems formalizes the ideas of using multiple distinct theories which have been exploited in much applied work in various areas of Artificial Intelligence, e.g. computational linguistics[74, 73], planning [54] and multiagent systems[46, 53, 69, 17, 18] The ML systems based on beliefs contexts provide the expressibility of normal modal logics [40] and of the most common non normal modal logics [33] The goal of this chapter is to apply the idea of Local Models Semantics to provide a ....

Wilks, Y., and Ballim, A. Multiple agents and the heuristic ascription of belief. In Proc. of the 10th International Joint Conference on Artificial Intelligence (1987), pp. 118--124.

.... which are preoccupied with the communicating agents, their goals, plans and beliefs; such as Speech Act Theory (Austin, 1962; Searle, 1969) Rhetorical Structure Theory (rst) Mann and Thompson, 1987) or AI research on scripts (Lehnert, 1981; Schank and Abelson, 1977) and belief ascription (Wilks and Ballim, 1987). ffl b) theories on the tracking of the discourse history by means of identifying the focused items therein; e.g. Grosz, 1986; Hobbs, 1978; Reichman, 1985; Sidner, 1983; Webber, 1983) ffl c) theories of cohesion and coherence and how these are manifested on the surface of the text; e.g. ....

Y. Wilks and A. Ballim. Multiple Agents and the Heuristic Ascription of Belief. In Proceedings of the 10th International Joint Conference on Artificial Intelligence (IJCAI-87), pages 119--124, Milan, Italy, August 1987.

....in the domains of human computer communication, and interactive document systems. As a research fellow of the Computing Research Laboratory (NMSU, USA) he worked on projects of knowledge representation [BCdRF89] natural language processing [WFBH89] and agent modeling in dialogue [BW90, BWB91, WB87] As a researcher at the Institute Dalle Molle pour les Etudes Semantique et Cognitives (ISSCO) he worked on numerous projects related to dialogue and text processing, natural language processing and machine assisted translation. These include a project on using structural information in ....

Y. Wilks and A. Ballim. Multiple agents and the heuristic ascription of belief. In Proceedings of the 10th International Joint Conference on Artificial Inteligence, pages 118--124. Morgan Kaufmann, 1987.

No context found.

Y. Wilks and A. Ballim. Multiple agents and the heuristic ascription of belief. In Proc. of the 10th International Joint Conference on Artificial Intelligence, pages 118--124, 1987.

No context found.

Y. Wilks and A. Ballim. Multiple agents and the heuristic ascription of belief. In Proc. of the 10th International Joint Conference on Artificial Intelligence, pages 118--124, 1987.

No context found.

Wilks, Yorick, & Ballim, Afzal (1987), "Multiple Agents and the Heuristic Ascription of Belief," Proceedings of the 10th International Joint Conference on Artificial Intelligence (IJCAI-87, Milan) (Los Altos, CA: Morgan Kaufmann): 118--124.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC