The intensive use of memory to recall specific episodes from the pastrather than rules-should be the foundation of machine reasoning. CRAIG STANFILL and DAVID WALTZThe traditional assumption in artificial intelligence (Al) is that most expert knowledge is encoded in the form of rules. We consider the phenomenon of reasoning from memories of specific episodes, however, to be the foundation of an intelligent system, rather than an adjunct to some other reasoning method. This theory contrasts with much of the current work in similarity-based learning, which tacitly assumes that learning is equivalent to the automatic generation of rules, and differs from work on "explanation-based" and "case-based" reasoning in that it does not depend on having a strong domain model.With the development of new parallel architectures, specifically the Connection Machine" system, the operations necessary to implement this approach to reasoning have become sufficiently fast to allow experimentation. This article describes MBRtalk. an experimental memory-based reasoning system that has been implemented on the Connection Machine, as well as the application of memory-based reasoning to other domains. THE MEMORY-BASED REASONING HYPOTHESISAlthough we do not reject the necessity of other forms of reasoning, including those that are currently modeled by rules or by the induction of rules, we believe that the theory behind memory-based The Conneclion Macliinc is a registered trademark of Thinking Machines Corjiiiralion, (CJ 1986 ACM 0001-0782/86/1200-121H 7S(E Primitive Operations of Memory-Based ReasoningThe memory-based reasoning hypothesis has not been extensively studied in the past because von Neumann machines do not support it well.

No abstract

Abstract-Power companies can benefit from the use of knowledge discovery methods and statistical machine learning for preventive maintenance. We introduce a general process for transforming historical electrical grid data into models that aim to predict the risk of failures for components and systems. These models can be used directly by power companies to assist with prioritization of maintenance and repair work. Specialized versions of this process are used to produce 1) feeder failure rankings, 2) cable, joint, terminator and transformer rankings, 3) feeder MTBF (Mean Time Between Failure) estimates and 4) manhole events vulnerability rankings. The process in its most general form can handle diverse, noisy, sources that are historical (static), semi-real-time, or real-time, incorporates state-of-the-art machine learning algorithms for prioritization (supervised ranking or MTBF), and includes an evaluation of results via cross-validation and blind test. Above and beyond the ranked lists and MTBF estimates are business management interfaces that allow the prediction capability to be integrated directly into corporate planning and decision support; such interfaces rely on several important properties of our general modeling approach: that machine learning features are meaningful to domain experts, that the processing of data is transparent, and that prediction results are accurate enough to support sound decision making. We discuss the challenges in working with historical electrical grid data that were not designed for predictive purposes. The "rawness" of these data contrasts with the accuracy of the statistical models that can be obtained from the process; these models are sufficiently accurate to assist in maintaining New York City's electrical grid.Index Terms-applications of machine learning, electrical grid, smart grid, knowledge discovery, supervised ranking, computational sustainability, reliability !

No abstract

No abstract