twelve years after

In this survey, we review work in machine learning on methods for handling data sets containing large amounts of irrelevant information. We focus on two key issues: the problem of selecting relevant features, and the problem of selecting relevant examples. We describe the advances that have been made on these topics in both empirical and theoretical work in machine learning, and we present a general framework that we use to compare different methods. We close with some challenges for future work in this area.

This article describes a comprehensive approach to automatic theory revision. Given an imperfect theory, the approach combines explanation attempts for incorrectly classified examples in order to identify the failing portions of the theory. For each theory fault, correlated subsets of the examples are used to inductively generate a correction. Because the corrections are focused, they tend to preserve the structure of the original theory. Because the system starts with an approximate domain theory, in general fewer training examples are required to attain a given level of performance (classification accuracy) compared to a purely empirical system. The approach applies to classification systems employing a propositional Horn-clause theory. The system has been tested in a variety of application domains, and results are presented for problems in the domains of molecular biology and plant disease diagnosis. 1 INTRODUCTION 2 1 Introduction One of the most difficult problems in the develo...

Knowledge acquisition is a difficult, error-prone, and time-consuming task. The task of automatically improving an existing knowledge base using learning methods is addressed by the class of systems performing theory refinement. This paper presents a system, Forte (First-Order Revision of Theories from Examples), which refines first-order Horn-clause theories by integrating a variety of different revision techniques into a coherent whole. Forte uses these techniques within a hill-climbing framework, guided by a global heuristic. It identifies possible errors in the theory and calls on a library of operators to develop possible revisions. The best revision is implemented, and the process repeats until no further revisions are possible. Operators are drawn from a variety of sources, including propositional theory refinement, first-order induction, and inverse resolution. Forte is demonstrated in several domains, including logic programming and qualitative modelling.

This chapter describes a multistrategy system that employs independent modules for deductive, abductive, and inductive reasoning to revise an arbitrarily incorrect propositional Horn-clause domain theory to t a set of preclassi ed training instances. By combining such diverse methods, Either is able to handle a wider range of imperfect theories than other theory revision systems while guaranteeing that the revised theory will be consistent with the training data. Either has successfully revised two actual expert theories, one in molecular biology and one in plant pathology. The results con rm the hypothesis that using a multistrategy system to learn from both theory and data gives better results than using either theory or data alone. 1

This paper describes an integrated system for coordinating multiple rover behavior with the overall goal of collecting planetary surface data. The MultiRover Integrated Science Understanding System combines concepts from machine learning with planning and scheduling to perform autonomous scientic exploration by cooperating rovers. The integrated system utilizes a novel machine learning clustering component to analyze science data and direct new science activities. A planning and scheduling system is employed to generate rover plans for achieving science goals and to coordinate activities among rovers. We describe each of these components and discuss some of the key integration issues that arose during development and in- uenced both system design and performance. Introduction Landmark events have recently taken place in the areas of space exploration and planetary rovers. The Mars Pathnder mission was a major success, not only demonstrating the feasibility of sending...

Most research on computational scientific discovery has focused on developing an initial model, but an equally important task involves revising a model in response to new data. In this paper, we present an approach that represents candidate models as sets of quanti-tative processes and that treats revision as search through a model space which is guided by time-series observations and constrained by background knowledge cast as generic processes. We demonstrate our system’s ability on three different scientific domains and associated data sets. We also discuss its relation to other work on model revision and consider directions for additional research. 1

In this paper we explore the idea of reflection in the context of scientific exploration. How does an agent reflect upon its behavior in order to enable productive exploration? We outline an abstract cognitive architecture for combining reflection and exploration. To achieve this we present a language for modeling cognition: the Task-Method-Knowledge(TMK) language. We further present a computational model based on this language, ToRQUE2 (Griffith et al., 1997; Griffith, 1997). ToRQUE2 is a model of exploratory reasoning in the domain of scientific problem solving. We claim that the TMK language supports both reflection and exploration, and enables them to benefit from one another. Introduction One outstanding issue in scientific discovery is: how do scientists decide when to abandon one reasoning strategy to pursue another? Or from a cognitive perspective: how do we model a process for multi-strategy exploration? In our research we show evidence from a computational system...

In the process of knowledge acquisition, inductive learning and theory revision play important roles. Inductive learning is used to acquire new knowledge (theories) from training examples; and theory revision improves an initial theory with training examples. A theory preference criterion is critical in the processes of inductive learning and theory revision. A new system called knowar is developed by integrating inductive learning and theory revision. In addition, the theory preference criterion used in knowar is the combination of the MDL-based heuristic and the Laplace estimate. The system can be used to deal with complex problems. Empirical studies have confirmed that knowar leads to substantial improvement of a given initial theory in terms of its predictive accuracy. keywords: knowledge acquisition, inductive logic programming, theory revision, the MDL principle, the Laplace estimate, noisy data. 1 Introduction Knowledge acquisition includes theory formation and theory revisio...

Analogy is often believed to play an important role in reasoning underlying innovation and creativity. The ability to make analogies between distant situations or domains (i.e., cross-domain analogies) appears to be crucial for innovation and creativity. However, making cross-domain analogies often involves learning shared abstractions as well as reasoning mediated by the abstractions. We hypothesize that structure-behaviorfunction (SBF) models at different levels of abstraction provide the right knowledge to facilitate analogical reasoning, ranging from within-domain to cross-domain analogies. We call such analogical reasoning model-based analogy. A mental model is characterized by the types of information it captures such as causal, functional (teleological), and structural relations between the entities in a system or a situation. We represent device-specific models (i.e., models of specific designs) as SBF models and device-independent models (i.e., models of physical principles, ...