Abstract not found

Abstract. An important step in the solution of a target problem in case-based reasoning (CBR) is the retrieval of similar previous cases that can be used to solve the target problem. We review a selection of papers from the CBR literature on aspects of retrieval, such as approaches to the assessment of surface and structural similarity and techniques for automating the construction and maintenance of similarity measures. We also examine a number of retrieval techniques that have been developed to address the limitations of retrieval based purely on similarity. 1

. In this paper we present a probabilistic formalization of the instance-based learning approach. In our Bayesian framework, moving from the construction of an explicit hypothesis to a data-driven instancebased learning approach, is equivalent to averaging over all the (possibly infinitely many) individual models. The general Bayesian instance-based learning framework described in this paper can be applied with any set of assumptions defining a parametric model family, and to any discrete prediction task where the number of simultaneously predicted attributes is small, which includes for example all classification tasks prevalent in the machine learning literature. To illustrate the use of the suggested general framework in practice, we show how the approach can be implemented in the special case with the strong independence assumptions underlying the so called Naive Bayes classifier. The resulting Bayesian instance-based classifier is validated empirically with public domain data sets...

A memory-based learning system is an extended memory management system that decomposes the input space either statically or dynamically into subregions for the purpose of storing and retrieving functional information. The main generalization techniques employed by memory-based learning systems are the nearest-neighbor search, space decomposition techniques, and clustering. Research on memory-based learning is still in its early stage. In particular, there are very few rigorous theoretical results regarding memory requirement, sample size, expected performance, and computational complexity. In this paper, we propose a model for memory-based learning and use it to analyze several methods--- ffl-covering, hashing, clustering, tree-structured clustering, and receptive-fields---for learning smooth functions. The sample size and system complexity are derived for each method. Our model is built upon the generalized PAC learning model of Haussler (Haussler, 1989) and is closely related to the method of vector quantization in data compression. Our main result is that we can build memory-based learning systems using new clustering storage in typical situations.

Abstract|We describe an unsupervised on-line method for learning of manipulative actions that allows a robot to push an object connected to it with a rotational point contact to a desired point in image-space. By observing the results of its actions on the object's orientation in imagespace, the system forms a predictive forward empirical model. This acquired model is used on-line for manipulation planning and control as it improves. Rather than explicitly inverting the forward model to achieve trajectory control, a stochastic action selection technique [Moore, 1990] is used to select the most informative and promising actions, thereby integrating active perception and learning by combining on-line improvement, task-directed exploration, and model exploitation. Simulation and experimental results of the approach are presented. I.

Model-based reinforcement learning, in which a model of the environment's dynamics is learned and used to supplement direct learning from experience, has been proposed as a general approach to learning and planning. We present experiments with this idea in which the model of the environment's dynamics is both approximate and learned online. These experiments involve the Mountain Car task, which requires approximation of both value function and model because it has continuous state variables. Naive model use is susceptible to modelling errors and can impair the value function. We show that excessive model use performs worse than using no model at all. Hybrid methods can mitigate learning with inherently incorrect models. 1 Introduction Some of the most impressive successes of reinforcement learning so far have used extensive offline experience with a model or simulation of the task in order to attain a high level of performance (Tesauro 1992; Crites & Barto 1996; Zhang & Dietterich 19...

Recent research in machine learning has focused on supervised induction for simple classification and reinforcement learning for simple reactive behaviors. In the process, the field has become disconnected from AI's original goal of creating complete intelligent agents. In this paper, I review recent work on machine learning for planning, language, vision, and other topics that runs counter to this trend and thus holds interest for the broader AI research community. I also suggest some steps to encourage further research along these lines. Introduction A central goal of artificial intelligence has long been to construct a complete intelligent agent that can perceive its environment, generate plans, execute those plans, and communicate with other agents. The pursuit of this dream naturally led many researchers to focus on the component tasks of perception, planning, control, and natural language, or on generic issues that cut across these tasks, such as representation and search. Over ...

This paper considers the importance of exploration to game-playing programs which learn by playing against opponents. The central question is whether a learning program should play the move which offers the best chance of winning the present game, or if it should play the move which has the best chance of providing useful information for future games. An approach to addressing this question is developed using probability theory, and then implemented in two different learning methods. Initial experiments in the game of Go suggest that a program which takes exploration into account can learn better against a knowledgeable opponent than a program which does not. 1 Introduction One of the earliest aspirations of Artificial Intelligence was to develop computer game playing programs which could improve their play through experience, adapt their strategy to compete against a variety of opponents, and ultimately outplay their programmers. As in most learning problems, a program 1 Parts of t...

In this paper a learning framework to deal with restructurable control of a single-output dynamic plant is proposed. The central concept used to represent the restructurable behavior of the plant, and subsequently for the design of the framework, is the behavioral graph. The nodes of this graph correspond to possible local behaviors of the system while its edges model the switching scheme of the plant among its local behaviors. In the definition of this concept, General Dynamical System theory is used. The framework is able to learn the dynamics (models) of a reconfigurable system, select appropriate models, and ultimately control the plant according to given specifications. The framework design borrows concepts and techniques from the active fields of adaptive and learning control. The underlying ideas and the software prototype implementing the framework design are tested through a series of simulated experiments. The simulations demonstrate the feasibility of the approach for contro...

The problem of finding the set of most relevant cases from a given database, with respect to the decision making situation at hand, is frequently encountered in many realworld domains. In the case-based reasoning framework this task is commonly known as the case matching problem. Case matching is an important problem in several commercially significant application areas, such as industrial configuration and manufacturing problems. Earlier approaches to the case matching problem typically rely on some distance measure, e.g., the Euclidean distance, although there is no a priori guarantee that such measures really reflect the useful similarities and dissimilarities between the cases. In this paper we introduce a novel approach to the case matching problem based on Bayesian probability theory, and propose a Bayesian case matching measure for scoring the cases with respect to a given decision making situation. The Bayesian case matching score discussed is currently being applied in a real-...