Thedevelopmentofmultistrategylearningsystemsrequiresaclearunderstandingoftherolesandthe applicabilityconditionsofdifferentlearningstrategies.Tothisend,thischapterintroducesthe InferentialTheoryofLearning thatprovidesaconceptualframeworkforexplaininglogicalcapabilities oflearningstrategies,i.e.,their competence.Viewinglearningasaprocessofmodifyingthelearner's knowledgebyexploringthelearner'sexperience,thetheorypostulatesthatanysuchprocesscanbe describedasasearchina knowledgespace, which involvesthelearner'sexperience,piorknowledgeand the learninggoal .Thesearchoperatorsareinstantiationsof knowledgetransmutations, whichare genericpatternsofknowledgechange.Transmutationsmayemployanybasictypeofinference --- deduction,inductionoranalogy.Severalfundamentalknowledg etransmutationsaredescribedinanovel andgeneralway,suchasgeneralization,abstraction,explanationandsimilization,andtheircounterparts, specialization,concretion,predictionanddissimilization,respectively.Generalizationenlargesthe referenceset ofadescription(thesetofentitiesthatarebeingdescribed).Abstractionreducesthe amountofthedetailaboutthereferenceset.Explanationgeneratespremisesthatexplain(orimply)the givenpropertiesofthereferenceset.Similization transfersknowledgefromonereferencesettoasimilar referenceset.Usingconceptsofthetheory,a multistrategytask -adaptivelearning(MTL)methodology isoutlined,andillustratedbyanexample.MTLdynamicallyadaptsstrategiestothe learningtask , definedbytheinputinformation,learner'sbackgroundknowledge,andthelearninggoal. Thegoalof MTLresearchisto synergisticallyintegrateawiderangeofinferentiallearningstrategies,suchas empiricalgeneralization,constructiveinduction, deductivegeneralization,explanation,prediction, abstraction,andsimilization. Keywords: learningtheory,inferencetheory,multi...

Multistrategy task-adaptive learning (MTL) comprises a class of methods in which the learner determines by itself which strategy or combination of strategies is most appropriate for a given learning task defined by the learner's goal, the leamer's background knowledge (BK) and the input to the learning process. The paper presents a MTL method which is based on building a plausible justification that the learner's input is a consequence of its BK. The method assumes a general learning goal of deriving any useful knowledge from a given input and integrates dynamically a whole range of learning sategies. It also behaves as a singlestrategy method when the relationship between the input and the BK satisfies the requirements of the single-strategy method, and the general learning goal of the MTL method is specialized to the goal of the single-strategy method.

This paper presents a learning-based approach to the automation of knowledge acquisition for expert systems. An expert system is viewed as an explicit mooel of a human expert's competence and perfonnance. We distinguish three phases in the development of such a model. The fIrst one consists of defIning a framework for the mooel, in terms of a knowledge representation formalism and an associated problem solving methoo. The second phase consists of defIning a preliminary mooel that describes the basic concepts of the expertise domain. The last phase consists of incrementally extending and improving the domain model through learning from the human expert. The paper describes the learning system NeoDISCIPLE which illustrates the usefulness of six principles for automating the knowledge acquisition process: expert system building as a threephase mooeling of human expertise, understanding-based knowledge extension, knowledge acquisition through multistrategy learning, consistency-driven concept fonnation and refinement, closed-loop learning, and cooperation between the human expert and the learning system.

Abstract not found

The development of multistrategy learning systems should be based on a clear understanding of the roles and the applicability conditions of different learning strategies. To this end, this report introduces the Inferential Theory of Learning that provides a conceptual framework for analyzing and explaining logical capabilities of learning strategies, that is, their competence. Viewing learning as a process of modifying the learner's knowledge by exploring the learner's experience, the theory postulates that any such process can be described as a search in a knowledge space, defined by the employed knowledge representation. The search operators are instantiations of knowledge transmutations, which are generic patterns of knowledge change. Transmutations may use any type of inference--deduction, induction or analogy. Several fundamental transmutations are presented in a novel and general way. These include generalization and specialization, abduction and prediction, abstraction and concretion, and similization and dissimilization. Generalization and specialization change the reference set of a description (the set of entities being described or referred to). Abstractions and concretions change the level of detail in describing the reference set. Explanations and predictions derive additional knowledge about the reference set (explanatory or predictive). Similizafions and dissimilizations hypothesize knowledge about a reference set based on its similarity or dissimilarity with another reference set. Using concepts of the theory, a multistrategy task-adaptive learning (MTL) methodology is outlined and illustrated by an example. MTL dynamically adapts strategies to the learning task, defined by the input information, learner's background knowledge, and the learning goal. ...