Truth maintenance is a collection of techniques for doing belief revision. A truth maintenance system's task is to maintain a set of beliefs in such a way that they are not known to be contradictory and no belief is kept without a reason. Truth maintenance systems were introduced in the late seventies by Jon Doyle and in the last five years there has been an explosion of interest in this kind of systems. In this paper we present an annotated bibliography to the literature of truth maintenance systems, grouping the works referenced according to several classifications.

Clancey troubleshooting tutor for only 20 The cognitive modeling group provided would like to see alternative cognitive hours gained a proficiency equivalent strong advocacy for the use of models available within a system to that of trainees with 40 months cognitive modeling in building these rather than a single "correct" model (almost 4 years) on-the-job training systems. They argued for increased used to justify instruction.

Molecular biology is emerging as an important domain for artificial intelligence research. The advantages of biology for design and testing of AI systems include large amounts of available online data, significant (but incomplete) background knowledge, a wide variety of problems commensurate with AI technologies, clear standards of success, cooperative domain experts, non-military basic research support and percieved potential for practical (and profitable) applications. These considerations have motivated a growing group of researchers to pursue both basic and applied AI work in the domain. More than seventy-five researchers working on these problems gathered at Stanford for a AAAI sponsored symposium on the topic. This article provides a description of much of the work presented at the meeting, and fills in the basic biology background necessary to place it in context.

These algorithms are also related to time-independent fonnalisms.

We present an application of back-propagation networks to handwritten digitrecognition. Minimal preprocessing of the data was required, but architecture of the network was highly constrained and specifically designed for the task. The input of the network consists of normalized images of isolated digits. The method has 1 % error rate and about a 9% reject rate on zipcode digits provided by the U.S. Postal Service. 1 INTRODUCTION The main point of this paper is to show that large back-propagation (BP) networks canbe applied to real image-recognition problems without a large, complex preprocessing stage requiring detailed engineering. Unlike most previous work on the subject (Denker et al., 1989), the learning network is directly fed with images, rather than feature vectors, thus demonstrating the ability of BP networks to deal with large amounts of low level information. Previous work performed on simple digit images (Le Cun, 1989) showed that the architecture of the network strongly influences the network's generalization ability. Good generalization can only be obtained by designing a network architecture that contains a certain amount of a priori knowledge about the problem. The basic design principleis to minimize the number of free parameters that must be determined by the learning algorithm, without overly reducing the computational power of the network.

A new concept for unsupervised learning based upon examples introduced tothe neural network is proposed. Each example is considered as an interpolation node of the velocity field in the phase space. The velocities at these nodes are selected such that all the streamlines converge to an attracting set imbedded in the subspace occupied by the cluster of examples. The synaptic interconnections are found from learning procedure providing selected field. The theory is illustrated by examples. This paper is devoted to development of a new concept for unsupervised learning based upon examples introduced to an artificial neural network.

Pentti Kanerva Research Institute for Advanced Computer Science Mail Stop 230-5, NASA Ames Research Center Moffett Field, California 94035 ABSTRACT Contour maps provide a general method for recognizing two-dimensional shapes. All but blank images give rise to such maps, and people are good at recognizing objects and shapes from them. The maps are encoded easily in long feature vectors that are suitable for recognition by an associative memory. These properties of contour maps suggest a role for them in early visual perception. The prevalence of direction-sensitive neurons in the visual cortex of mammals supports this view.

A rigorous analysis on the finite precision computational)Spects of neural network as a pattern classifier via a probabilistic approach is presented.

A methodology for faster supervised learning in dynamical nonlinear neuralnetworks is presented. It exploits the concept of adjoint operntors to enable computation of changes in the network's response dueto perturbations in all system parameters, using the solution of a single set of appropriately constructed linear equations. The lower bound on speedup per learning iteration over conventional methodsfor calculating the neuromorphic energy gradient is O(N2), where N is the number of neurons in the network. 1 INTRODUCTION The biggest promise of artifcial neural networks as computational tools lies in the hope that they will enable fast processing and synthesis of complex information patterns. In particular, considerable efforts have recently been devoted to the formulation ofefficent methodologies for learning (e.g., Rumelhart et al., 1986; Pineda, 1988; Pearlmutter, 1989; Williams and Zipser, 1989; Barhen, Gulati and Zak, 1989). The development of learning algorithms is generally based upon the minimization of a neuromorphic energy function.

Kanerva's sparse distributed memory (SDM) is an associative-memory modelbased on the mathematical properties of high-dimensional binary address spaces. Holland's genetic algorithms are a search technique forhigh-dimensional spaces inspired by evolutionary processes of DNA. "Genetic Memory" is a hybrid of the above two systems, in which the memory uses a genetic algorithm to dynamically reconfigure itsphysical storage locations to reflect correlations between the stored addresses and data. For example, when presented with raw weather station data, the Genetic Memory discovers specific features inthe weather data which correlate well with upcoming rain, and reconfigures the memory to utilize this information effectively. This architecture is designed to maximize the ability of the system to scale-up to handle real-world problems.