Explanation-Based Generalization: A Unifying View

By Tom Michael Mitchell, Richard M. Keller, Smadar T. Kedar-Cabelli. 1986

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistical Learning Theory

Citations in the corpus

Joint Concept Learning and Semantic Parsing from Natural Language Explanations (2017, background): “Earlier work on Explanation-based learning (Mitchell et al., 1986; DeJong and Mooney, 1986) Semantic interpretation of language has been explored in diverse domains.”
Deep learning in neural networks: An overview (2014, methodology): “A hybrid approach (Shavlik and Towell, 1989; Towell and Shavlik, 1994) initializes a potentially deep FNN through a domain theory in propositional logic, which may be acquired through explanation-based learning (Mitchell et al., 1986; DeJong and Mooney, 1986; Minton et al., 1989).”
A historical perspective of explainable Artificial Intelligence (2020, background): “While this kind of explanation mainly accommodates the need of knowledge engineers to understand whether the system's reasoning is technically sound, it (or slight variants of it) can also be provided as an explanation to domain experts (Buchanan & Shortliffe, 1984; Mitchell et al., 1986).”
Programming by demonstration: dual-arm manipulation tasks for humanoid robots (2004, methodology): “For generalizing a single demonstration mainly explanation based methods are used [7], [8].”
Learning in the Presence of Concept Drift and Hidden Contexts (1996, background): “Their experiments had nothing to do with concept drift, but were motivated by the so-called utility problem in Explanation-Based Learning ( Mitchell et al., 1986 ), which is the problem that learned macro-operators or schemata, even if correct, are not always helpful, but may actually decrease the p”
Explanation-Based Learning: An Alternative View (1986, background): “uninstantiated versions of the general rules (an explanation structure (Mitchell et al., 1986)) and two independent substitution lists called SPECIFIC and GENERAL.”
A Review and Empirical Evaluation of Feature Weighting Methods for a Class of Lazy Learning Algorithms (1997, background): “They assign instance-speciic weights to discrete features using an explanation-based learning (EBL) approach (Mitchell et al., 1986).”
Informed Machine Learning – A Taxonomy and Survey of Integrating Prior Knowledge into Learning Systems (2019, background): “Logic rules have a long tradition in symbolic AI, in particular in expert systems based on rules [63], [64].”
Contextual Explanation Networks (2017, methodology): “Using explanations based on domain knowledge is known to improve generalization [28] and could be used as a powerful mechanism for solving complex downstream tasks such as program induction for solving algebraic word problems [29].”
Active Learning with Feedback on Features and Instances (2006, background): “The field of explanation-based learning, however, concerns itself on a deductive rather than an inductive learning task, using one instance and a given domain theory to generalize (Mitchell et al., 1986; DeJong and Mooney, 1986).”
Robot task planning and explanation in open and uncertain worlds (2015, background): “There is also a relationship to work in explanation-based learning [30].”
Acquiring Recursive and Iterative Concepts with Explanation-Based Learning (1990, background): “This question involves the relationship between the operationality and generality of acquired rules [DeJong & Mooney, 1986; Keller, 1988; Mitchell et al., 1986; Segre, 1987; Shavlik, DeJong, & Ross, 1987].”