In many machine learning applications, one has access, not only to training data, but also to some high-level a priori knowledge about the desired behavior ofthe system. For example, it is known in advance that the output of a character recognizer should be invariant with respect to small spatial distortionsof the input images (translations, rotations, scale changes, etcetera). We have implemented a scheme that allows a network to learn the derivative ofits outputs with respect to distortion operators of our choosing. This not only reduces the learning time and the amount of training data, but also provides a powerful language for specifying what generalizations we wish the network to perform. 1 INTRODUCTION In machine learning, one very often knows more about the function to be learned than just the training data. An interesting case is when certain directional derivatives ofthe desired function are known at certain points.

Because of the complexity of the application and the large amount of data, the problem cannot be solved by using a single method. The solution we propose is to build multimodules NNarchitectures where several NN cooperate together. Such system suffer from generic problem for whom we propose solutions. They allow to reach accurate performances for multi-valued function approximations and probability estimations. The results are compared with six other methods which have been used for this problem. We show that the methodology we have developed is general and can be used for a large variety of applications.

Accurate saccades require interaction between brainstem circuitry and the cerebeJJum. A model of this interaction is described, based on Kawato's principle of feedback-error-Iearning. In the model a part of the brainstem (the superior colliculus) acts as a simple feedback controJJer with no knowledge of initial eye position, and provides an error signal for the cerebeJJum to correct for eye-muscle nonIinearities. This teaches the cerebeJJum, modelled as a CMAC, to adjust appropriately the gain on the brainstem burst-generator's internal feedback loop and so alter the size of burst sent to the motoneurons. With direction-only errors the system rapidly learns to make accurate horizontal eye movements from any starting position, and adapts realistically to subsequent simulated eye-muscle weakening or displacement of the saccadic target.

Many real-valued connectionist approaches to learning control realize exploration by randomness inaction selection. This might be disadvantageous when costs are assigned to "negative experiences" . The basic idea presented in this paper is to make an agent explore unknown regions in a more directed manner. This is achieved by a so-called competence map, which is trained to predict the controller's accuracy, and is used for guiding exploration. Based on this, a bistable system enables smoothly switching attention between two behaviors - exploration and exploitation - depending on expected costsand knowledge gain. The appropriateness of this method is demonstrated by a simple robot navigation task.

After being trained on some dozen Bach chorales using error backpropagation, the system is capable of producing four-part chorales in the style of J .s.Bach, given a one-part melody. Our system solves a musical real-world problem on a performance level appropriate for musical practice. HARMONET's power is based on (a) a new coding scheme capturing musically relevant information and (b) the integration of backpropagation and symbolic algorithms in a hierarchical system, combining theadvantages of both. 1 INTRODUCTION Neural approaches to music processing have been previously proposed (Lischka, 1989) and implemented (Mozer, 1991)(Todd, 1989). The promise neural networks offer is that they may shed some light on an aspect of human creativity that doesn't seem to be describable in terms of symbols and rules. Ultimately what music is (or isn't) lies in the eye (or ear) of the beholder.

We use connectionist modeling to develop an analysis of stress systems in terms of ease of learnability. In traditional linguistic analyses, learnability arguments determine default parameter settings based on the feasibilty of logically deducing correct settings from an initial state. Our approach provides an empirical alternative tosuch arguments. Based on perceptron learning experiments using data from nineteen human languages, we develop a novel characterization of stress patterns in terms of six parameters. These provide both a partial description of the stress pattern itself and a prediction of its learnability, without invoking abstract theoretical constructs such as metrical feet. This work demonstrates that machine learningmethods can provide a fresh approach to understanding linguistic phenomena.

The focused gamma network is proposed as one of the possible implementations of the gamma neural model. The focused gamma network is compared with the focused backpropagation network and TDNN for a time series prediction problem, and with ADALINE in a system identification problem.

I met Allen for the first time when I came for a two semester long visit to Carnegie Mellon University in 1968. This encounter was a distinct factor in my later decision to join the faculty at Carnegie Mellon University.

This technology was developed with funding from the National Sony, the Japanese consumer electronics Science Foundation. Working with experts from Armco Steel (Middletown, company, has developed OH), Carnegie Group developed a prototype system to diagnose an intelligent system to improve chatter in a coldrolling mill. In the and consulting company, has developed a PCbased virtual reality system company's semiconductor group, to provide financial planners a visual metaphor for viewing large The system allows the user to "fly" over the The expert system is installed in Meiji's Tokyo service two-thirds. With Domain Dynamics Ltd. (Windsor, England) has developed a PCbased the system, technical support neural network application to automate the recognition of data from the Currently available in days to solve with a text retrieval the form of two circuit boards, TESPAR (which stands for Time Encoded system now take just afew minutes. Signal Processing and Recognition) is capable of being converted to a single piece of silicon.

Another issue concerned The survey of 80 KBS developers in Knowledge-Based Systems, Marc Ayel structural validation of KBS, given financial domains by Daniel O'Leary and Jean-Pierre Laurent, eds., John that the architecture of these systems opens the collection and raises some Wiley and Sons, Chichester, England, (having separate knowledge base and interesting points. VVT is revealed to 1991, 219 pp., $49.95, ISBN 0-471- inference engine components) was be a significant concern, with developers 93018-0 (paper). Testing his volume contains a selection First European Workshop on studies were launched to determine with real and contrived test cases was Verification, Validation, and Test of the applicability of software engineering found to account for about half the Knowledge-Based Systems, held evaluation techniques to overall VVT effort on average, with during the 1990 European Conference KBSs and to develop new techniques direct inspection of the knowledge on Artificial Intelligence (ECAI specific to KBSs. Several such studies base accounting for another 30 percent 90) in Stockholm, Sweden. In reviewing were initiated by organizations that in the survey.