This paper describes the integration of analogical reasoning into general problem solving as a method of learning at the strategy level to solve problems more effectively. The method based on derivational analogy has been fully implemented in prodigy/analogy and proven empirically to be amenable to scaling up both in terms of domain and problem complexity. prodigy/analogy addresses a set of challenging problems, namely: how to accumulate episodic problem solving experience, cases, how to define and decide when two problem solving situations are similar, how to organize a large library of planning cases so that it may be efficiently retrieved, and finally how to successfully transfer chains of problem solving decisions from past experience to new problem solving situations when only a partial match exists among corresponding problems. The paper discusses the generation and replay of the problem solving cases and we illustrate the algorithms with examples. We present briefly the librar...

A novel architecture is presented for combining rule-based and case-based reasoning. The central idea is to apply the rules to a target problem to get a first approximation to the answer; but if the problem is judged to be compellingly similar to a known exception of the rules in any aspect of its behavior, then that aspect is modelled after the exception rather than the rules. The architecture is implemented for the full-scale task of pronouncing surnames. Preliminary results suggest that the system performs almost as well as the best commercial systems. However, of more interest than the absolute performance of the system is the result that this performance was better than what could have been achieved with the rules alone. This illustrates the capacity of the architecture to improve on the rule-based system it starts with. The results also demonstrate a beneficial interaction in the system, in that improving the rules speeds up the case-based component. 1 Introduction One strategy...

. Retrieval of cases is one important step within the case-based reasoning paradigm. We propose an improvement of this stage in the process model for finding most similar cases with an average effort of O[log2n], n number of cases. The basic idea of the algorithm is to use the heterogeneity of the search space for a density-based structuring and to employ this precomputed structure, a k-d tree, for efficient case retrieval according to a given similarity measure sim. In addition to illustrating the basic idea, we present the experimental results of a comparison of four different k-d tree generating strategies as well as introduce the notion of virtual bounds as a new one that significantly reduces the retrieval effort from a more pragmatic perspective. The presented approach is fully implemented within the (Patdex) system, a case-based reasoning system for diagnostic applications in engineering domains. 1 Introduction Retrieval of sufficiently similar cases is one of the main points ...

This article describes how a reasoner can improve its understanding of an incompletely understood domain through the application of what it already knows to novel problems in that domain. Casebased reasoning is the process of using past experiences stored in the reasoner's memory to understand novel situations or solve novel problems. However, this process assumes that past experiences are well understood and provide good "lessons" to be used for future situations. This assumption is usually false when one is learning about a novel domain, since situations encountered previously in this domain might not have been understood completely. Furthermore, the reasoner may not even have a case that adequately deals with the new situation, or may not be able to access the case using existing indices. We present a theory of incremental learning based on the revision of previously existing case knowledge in response to experiences in such situations. The theory has been implemented in a case-base...

In this paper, we address the problem of case-based learning in the presence of irrelevant features. We review previous work on attribute selection and present a new algorithm, Oblivion, that carries out greedy pruning of oblivious decision trees, which effectively store a set of abstract cases in memory. We hypothesize that this approach will efficiently identify relevant features even when they interact, as in parity concepts. We report experimental results on artificial domains that support this hypothesis, and experiments with natural domains that show improvement in some cases but not others. In closing, we discuss the implications of our experiments, consider additional work on irrelevant features, and outline some directions for future research.

: Learning algorithms differ in the degree to which they process their inputs prior to their use in performance tasks. Many algorithms eagerly compile input samples and use only the compilations to make decisions. Others are lazy: they perform less precompilation and use the input samples to guide decision making. The performance of many lazy learners significantly degrades when samples are defined by features containing little or misleading information. Distinguishing feature relevance is a critical issue for these algorithms, and many solutions have been developed that assign weights to features. This chapter introduces a categorization framework for feature weighting approaches used in lazy similarity learners and briefly surveys some examples in each category. 1.1 INTRODUCTION Lazy learning algorithms are machine learning algorithms (Mitchell, 1997) that are welcome members of procrastinators anonymous. Purely lazy learners typically display the following characteristics (Aha, 19...

Introspective reasoning about a system's own reasoning processes can form the basis for learning to refine those reasoning processes. The ROBBIE 1 system uses introspective reasoning to monitor the retrieval process of a case-based planner to detect retrieval of inappropriate cases. When retrieval problems are detected, the source of the problems is explained and the explanations are used to determine new indices to use during future case retrieval. The goal of ROBBIE's learning is to increase its ability to focus retrieval on relevant cases, with the aim of simultaneously decreasing the number of candidates to consider and increasing the likelihood that the system will be able to successfully adapt the retrieved cases to fit the current situation. We evaluate the benefits of the approach in light of empirical results examining the effects of index learning in the ROBBIE system. 1 Introduction A number of studies have examined the use of metareasoning to control the application of s...

Case-Based Planning (CBP) provides a way of scaling up domain-independent planning to solve large problems in complex domains. It replaces the detailed and lengthy search for a solution with the retrieval and adaptation of previous planning experiences. In general, CBP has been demonstrated to improve performance over generative (from-scratch) planning. However, the performance improvements it provides are dependent on adequate judgements as to problem similarity. In particular, although CBP may substantially reduce planning effort overall, it is subject to a mis-retrieval problem. The success of CBP depends on these retrieval errors being relatively rare. This paper describes the design and implementation of a replay framework for the case-based planner dersnlp+ebl. dersnlp +ebl extends current CBP methodology by incorporating explanation-based learning techniques that allow it to explain and learn from the retrieval failures it encounters. These techniques are used to refine judgemen...

. Case-based problem solving can be significantly improved by applying domain knowledge (in opposition to problem solving knowledge) , which can be acquired with reasonable effort, to derive explanations of the correctness of a case. Such explanations, constructed on several levels of abstraction, can be employed as the basis for similarity assessment as well as for adaptation by solution refinement. The general approach for explanation-based similarity can be applied to different real world problem solving tasks such as diagnosis and planning in technical areas. This paper presents the general idea as well as the two specific, completely implemented realizations for a diagnosis and a planning task. 1 Introduction and Motivation The underlying principle of case-based reasoning is the idea to remember solutions to already known problems for their reuse during novel problem solving. The case which is most similar to the current problem is retrieved from a case base and its solution is m...

In this paper, we address an issue that arises when the background knowledge used by explanationbased learning is incorrect. In particular, we consider the problems that can be caused by a domain theory that may be overly specific. Under this condition, generalizations formed by explanation-based learning will make errors of omission when they are relied upon to make predictions or explanations. We describe a technique for detecting errors of omission, assigning blame for the error of omission to an inference rule in the domain theory, and revising the domain theory to accommodate new examples. 1