LT Revisited: Experimental Results of Applying Explanation-Based Learning to the Logic of Principia Mathematica

O'Rorke, Paul

doi:10.1016/b978-0-934613-41-5.50019-2

Cited by 21 publications

(15 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In particular, experiments have compared different generalization strategies (O'Rorke, 1989), protocols for when to apply learning and how to use learned knowledge (Mooney, 1989), methods for learning recursive concepts (Etzioni, 1990;Subramanian & Feldman, 1990), the effectiveness of abstraction hierarchies in problem solving (Knoblock, 1990), and whether learned knowledge should eventually be overwritten or discarded (Markovitch & Scott, 1988).…”

Section: Resource Limit Biasmentioning

confidence: 99%

“…A number of experimental evaluations of explanation-based learning (EBL) (DeJong & Mooney, 1986;Mitchell, et al, 1986) have appeared in the machine learning literature (Eskey & Zweben, 1990;Etzioni, 1990;Knoblock, 1990;Markovitch & Scott, 1988;Mooney, 1989;O'Rorke, 1989;Tambe & Newell, 1988). These studies measure the performance of a learning system against the performance of a similar nonlearning system.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A critical look at experimental evaluations of EBL

1991

View full text Add to dashboard Cite

Abstract. A number of experimental evaluations of explanation-based learning (EBL) have been reported in the literature on machine learning. A close examination of the design of these experiments reveals certain methodological problems that could affect the conclusions drawn from the experiments. This article analyzes some of the more common methodological difficulties, and illustrates them using selected previous studies.

show abstract

Section: Resource Limit Biasmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A critical look at experimental evaluations of EBL

1991

View full text Add to dashboard Cite

show abstract

“…We agree with Davis that form alone is not the source of speed• EBL has been demonstrated to provide speedup in numerous problem-solving situations (for example, see O'Rorke, 1989). But there are factors that diminish the improvement offered by EBL--for example, the utility problem (Minton, 1988)• If too many useless rules are learned, EBL may degrade problem-solving performance instead of improving it.…”

Section: Knowledge Compilationmentioning

confidence: 53%

Explanation-based learning for diagnosis

Fattah

O'Rorke

1993

Mach Learn

Self Cite

View full text Add to dashboard Cite

Abstract. We present explanation-based learning (EBL) methods aimed at improving the performance of diagnosis systems integrating associational and model-based components. We consider multiple-fault model-based diagnosis (MBD) systems and describe two learning architectures. One, EBLIA, is a method for "learning in advance." The other, EBL(p), is a method for "learning while doing." EBLIA precompiles models into associations and relies only on the associations during diagnosis. EBL(p) performs compilation during diagnosis whenever reliance on previously learned associational rules results in unsatisfactory performance--as defined by a given performance threshold p. We present results of empirical studies comparing MBD without learning versus EBLIA and EBL(p). The main conclusions are as follows. EBLIA is superior when it is feasible, but it is not feasible for large devices. EBL(p) can speed-up MBD and scale-up to larger devices in situations where perfect accuracy is not required.

show abstract

“…EBL has been demonstrated to provide speedup in numerous problem-solving situations (for example, see O'Rorke, 1989). But there are factors that diminish the improvement offered by EBL-for example, the utility problem (Minton, 1988).…”

Section: Knowledge Compilationmentioning

confidence: 99%

Untitled

Fattah

O'Rorke

1993

Machine Learning

Self Cite

View full text Add to dashboard Cite

Abstract. We present explanation-based learning (EBL) methods aimed at improving the performance of diagnosis systems integrating associational and model-based components. We consider multiple-fault model-based diagnosis (MBD) systems and describe two learning architectures. One, EBLIA, is a method for "learning in advance." The other, EBL(p), is a method for "learning while doing." EBL IA precompiles models into associations and relies only on the associations during diagnosis. EBL(p) performs compilation during diagnosis whenever reliance on previously learned associational rules results in unsatisfactory performance-as defined by a given performance threshold p. We present results of empirical studies comparing MBD without learning versus EBL IA and EBL(p). The main conclusions are as follows. EBL IA is superior when it is feasible, but it is not feasible for large devices. EBL(p) can speed-up MBD and scale-up to larger devices in situations where perfect accuracy is not required.

show abstract

LT Revisited: Experimental Results of Applying Explanation-Based Learning to the Logic of Principia Mathematica

Cited by 21 publications

References 0 publications

A critical look at experimental evaluations of EBL

A critical look at experimental evaluations of EBL

Explanation-based learning for diagnosis

Untitled

Contact Info

Product

Resources

About