R. Oehlmann, D. Sleeman, and P. Edwards. Learning Plan Transformations from Self--Questions : A Memory--Based Approach. In Proceedings of the Eleventh National Conference on Artificial Intelligence, pages 520--525, Washington D.C., U.S.A., July 11 -- 15 1993. American Association for Artificial Intelligence.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....NONLIN [Tat77a, Tat77b] which allows the planning systems to annotate plans being created with information about the dependency structure between operators in the completed plan. This information can also be used to guide retrieval, reuse, and re planning. R. Oehlmann, D. Sleeman, and P. Edwards [OSE92, OSE93] suggest that in addition to the standard features of plan modification, execution, and repair, a planning system should not only be able to transform plans based on knowledge about actions, but also be able to learn the transformation strategies in order to apply to other plans in future use. ....

R. Oehlmann, D. Sleeman, and P. Edwards. Learning Plan Transformations from Self--Questions : A Memory--Based Approach. In Proceedings of the Eleventh National Conference on Artificial Intelligence, pages 520--525, Washington D.C., U.S.A., July 11 -- 15 1993. American Association for Artificial Intelligence.

....identify whether an operator is guaranteed to produce a correct result. Case based planning provides an potential solution to these problems. Traditional case based planning programs such as [2] have focused on planning in domains of physical action rather than in mental domains. More recent work [4, 8, 9] which has addressed mental domains has focused largely on the specific issue of using a meta level CBR process to develop new adaptation strategies on top of more traditional CBR. In contrast, we are interested in how case based planning can be extended into the mental domain of learning. Of ....

R. Oehlmann, D. Sleeman, and P. Edwards, Learning plan transformations from selfquestions: A memory-based approach. In Proceedings of the 11th National Conference on Artificial Intelligence, pp. 520-525, Cambridge, MA: AAAI-Press, 1993.

.... Supporting adaptation with introspective reasoning: Introspective reasoning about the adaptation process can be used to guide adaptation decisions and carry out adaptations and the search for needed information in a more flexible way (Leake, 1993a; Leake, 1995c; Leake, Kinley Wilson, Chapter 11; Oehlmann, 1993; Oehlmann, 1995) ffl Combining rules and cases for adaptation learning: Another new direction based on introspective reasoning is to combine rule based and case based adaptation, using reasoning from general heuristics when necessary, but whenever possible reusing more specific information from ....

Oehlmann, R.; Sleeman, D.; and Edwards, P. 1993. Learning plan transformations from selfquestions: A memory-based approach. In Proceedings of the Eleventh National Conference on Artificial Intelligence, 520--525. Menlo Park, CA: AAAI Press.

....strategies are higher level plans which organise the execution of single question and answer plans to generate questions in a particular sequence. The same basic plan structure used for question and answer plans has been employed for experimentation plans, although the index vocabulary differs (Oehlmann, Sleeman, Edwards, 1993). Experimentation (plan: name identify effect :collector What does the PIN WHEEL turn :collector1 :collector2 :collector3 :bindings ( focus object1 pin wheel) question goals (identify motion effect neighbouring object) question strategy goals (consequence checking) recovery goals ....

....not move. The source model can not be applied to the original electric circuit, because the switch and the plain pipe are not sufficiently similar. Therefore, the source plan is transformed into a new source plan able to generate a new source case which is sufficiently similar to the target case (Oehlmann, Sleeman, Edwards 1993) During plan transformation, IULIAN replaces the step which refers to the insertion of a plain pipe by a step which refers to the insertion of a valve. The valve is more similar to the switch in the target domain, because both components are used to pursue the goal select:flow interruption ....

[Article contains additional citation context not shown here]

Oehlmann, R., Sleeman, D., & Edwards, P. (1993). Learning Plan Transformations from SelfQuestions: A Memory-Based Approach. Proceedings of the 11th National Conference on Artificial Intelligence, (pp. 520-525). Cambridge, MA: AAAI-Press.

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