Results 1 
7 of
7
Nonmonotonic Reasoning, Preferential Models and Cumulative Logics
, 1990
"... Many systems that exhibit nonmonotonic behavior have been described and studied already in the literature. The general notion of nonmonotonic reasoning, though, has almost always been described only negatively, by the property it does not enjoy, i.e. monotonicity. We study here general patterns of ..."
Abstract

Cited by 540 (13 self)
 Add to MetaCart
Many systems that exhibit nonmonotonic behavior have been described and studied already in the literature. The general notion of nonmonotonic reasoning, though, has almost always been described only negatively, by the property it does not enjoy, i.e. monotonicity. We study here general patterns of nonmonotonic reasoning and try to isolate properties that could help us map the field of nonmonotonic reasoning by reference to positive properties. We concentrate on a number of families of nonmonotonic consequence relations, defined in the style of Gentzen [13]. Both prooftheoretic and semantic points of view are developed in parallel. The former point of view was pioneered by D. Gabbay in [10], while the latter has been advocated by Y. Shoham in [38]. Five such families are defined and characterized by representation theorems, relating the two points of view. One of the families of interest, that of preferential relations, turns out to have been studied by E. Adams in [2]. The pr...
Courteous Logic Programs: Prioritized Conflict Handling For Rules
, 1997
"... We define courteous logic programs, an expressive superclass of general logic programs, for the acyclic case. Courteous LP's feature not only classical negation as in extended LP's (Gelfond & Lifschitz), but also prioritized conflict handling. We show courteous LP's always have a consistent and uniq ..."
Abstract

Cited by 31 (8 self)
 Add to MetaCart
We define courteous logic programs, an expressive superclass of general logic programs, for the acyclic case. Courteous LP's feature not only classical negation as in extended LP's (Gelfond & Lifschitz), but also prioritized conflict handling. We show courteous LP's always have a consistent and unique answer set, which can be computed in O(m ) time, where m is the size of the groundinstantiated program, as compared to O(m) time for general LP's. Courteous LP's provide a method to resolve conflicts that arise in authoring (specifying), updating, and merging. This is especially useful for creation of rulebased intelligent agents by nontechnical authors, e.g., for commercial applications such as personalized information filtering and workflow. Current work includes: implementing courteous LP's for such applications, in IBM's RAISE system; generalizing expressively, e.g., to permit recursion; and developing methods for interactive acquisition of rules, e.g., conflict analysis and interagent communication.
Learning to Reason: The NonMonotonic Case
, 1995
"... We suggest a new approach for the study of the nonmonotonicity of human commonsense reasoning. The two main premises that underlie this work are that commonsense reasoning is an inductive phenomenon, and that missing information in the interaction of the agent with the environment may be as informat ..."
Abstract

Cited by 15 (8 self)
 Add to MetaCart
We suggest a new approach for the study of the nonmonotonicity of human commonsense reasoning. The two main premises that underlie this work are that commonsense reasoning is an inductive phenomenon, and that missing information in the interaction of the agent with the environment may be as informative for future interactions as observed information. This intuition is formalized and the problem of reasoning from incomplete information is presented as a problem of learning attribute functions over a generalized domain. We consider examples that illustrate various aspects of the nonmonotonic reasoning phenomena, which have been used over the years as "benchmarks" for various formalisms, and translate them into Learning to Reason problems. We demonstrate that these have concise representations over the generalized domain and prove that these representations can be learned efficiently. The framework developed suggests an "operational " approach to studying reasoning that is nevertheless ...
Prioritizing Multiple, Contradictory Sources in CommonSense Learning by Being Told; or, AdviceTaker Meets Bureaucracy
, 1995
"... An important commonsense capability is to learn by "taking advice", or "being told", from a variety of sources of information: messages from various other agents, reading various texts, etc.. Such learned or "assimilated" knowledge provides an important basis for one's actions. Yet a basic feature ..."
Abstract

Cited by 8 (5 self)
 Add to MetaCart
An important commonsense capability is to learn by "taking advice", or "being told", from a variety of sources of information: messages from various other agents, reading various texts, etc.. Such learned or "assimilated" knowledge provides an important basis for one's actions. Yet a basic feature of life is that one cannot believe everything that one is told. Not only may advice be incorrect; e.g., it may be contradicted by direct experience. Worse, different sources of advice information may contradict each other, or even themselves. How is an advicetaking agent to maintain a consistent, yet usefully actionable, set of beliefs, then? Here, we address this problem, which has not received much previous attention in the AI literature. We illustrate, with an actual example, how it arises in the domain of bureaucracy . A common human strategy is to treat advice as working belief, which can be overridden by other advice and direct experience. This suggests the first element of our approa...
Practical Prioritized Defaults via Logic Programs
, 1996
"... We define courteous logic programs, a variant of extended logic programs. Courteous logic programs constitute a kind of prioritized default reasoning that is computationally tractable, fairly powerful expressively, and commercially applicable to the creation of rulebased intelligent information age ..."
Abstract

Cited by 2 (1 self)
 Add to MetaCart
We define courteous logic programs, a variant of extended logic programs. Courteous logic programs constitute a kind of prioritized default reasoning that is computationally tractable, fairly powerful expressively, and commercially applicable to the creation of rulebased intelligent information agents. Priorities are, in part, specified explicitly: within the locale of each predicate's "definition ", i.e., between rules mentioning the same predicate in their consequent. (Recall that extended logic programs represent classical negation as well as negationasfailure. Hence a rule having consequent p may conflict with another rule having consequent :p.) In addition, priorities are, in part, specified implicitly: based on inferential dependency depth between the different predicate locales, as in the usual semantics for negation as failure in stratified logic programs. Syntactically, a courteous logic program is an extended logic program augmented with a prioritization subprogram. In th...
Zlog: Applying SystemZ
"... We present Zlog  a practical system that employs the systemZ[14] semantics. Zlog incurs polynomial cost for compilation and entailment in the horn and qhorn[3] cases. Zlog's complexity is intractable in the unrestricted case  but intractable in the number of defaults that cause the violat ..."
Abstract
 Add to MetaCart
We present Zlog  a practical system that employs the systemZ[14] semantics. Zlog incurs polynomial cost for compilation and entailment in the horn and qhorn[3] cases. Zlog's complexity is intractable in the unrestricted case  but intractable in the number of defaults that cause the violation of the qhorn property.
Defeasible and Pointwise Prioritization
, 1995
"... We suggest why, and show how, to represent defeasible reasoning about prioritizationtype precedence. We define Defeasible Axiomatized Policy (DAP) circumscription: it is the first formalism to express defeasible prioritization. DAP circumscription can represent one or more (generally, a finite ref ..."
Abstract
 Add to MetaCart
We suggest why, and show how, to represent defeasible reasoning about prioritizationtype precedence. We define Defeasible Axiomatized Policy (DAP) circumscription: it is the first formalism to express defeasible prioritization. DAP circumscription can represent one or more (generally, a finite reflective tower) of metalevels of such reasoning, without resorting to a more powerful logical language. We argue for the usefulness, and analyze the expressive significance, of this representational generalization. We show that it can often be achieved with only a modest increase in the mathematical complexity of inference: DAP circumscription often reduces to a series of prioritized predicate circumscriptions, for which inference procedures are currently available. DAP circumscription also offers an improved approach to pointwise prioritization and circumscription, even in the basic, monotonic case of reasoning about prioritization. We observe that unsatisfiability and representational awkw...