This tendency to preserve information about the path is not a characteristic of traditional finite-state automata. ENCODING PATH INFORMATION In a different set of experiments, we asked whether the SRN could learn to use the infonnation about the path that is encoded in the hidden units' patterns of activation. In one of these experiments, we tested whether the network could master length constraints. When strings generated from the small finite-state grammar may only have a maximum of 8 letters, the prediction following the presentation of the same letter in position number six or seven may be different. For example, following the sequence'TSSSXXV', 'V' is the seventh letter and only another'V' would be a legal successor.

A variety of approaches to adaptive information processing have been developed by workers in disparate disciplines. These include the large body of literature on approximation and interpolation techniques (curve and surface fitting), the linear, real-time adaptive signal processing systems (such as the adaptive linear combiner and the Kalman filter), and most recently, the reincarnation of nonlinear neural network models such as the multilayer perceptron. Each of these methods has its strengths and weaknesses. The curve and surface fitting techniques are excellent for off-line data analysis, but are typically not formulated with real-time applications in mind. The linear techniques of adaptive signal processing and adaptive control are well-characterized, but are limited to applications for which linear descriptions are appropriate. Finally, neural network learning models such as back propagation have proven extremely versatile at learning a wide variety of nonlinear mappings, but tend to be very slow computationally and are not yet well characterized.

One att·empt at explaining human inferencing is that of spreading activat,ion, particularly in the st.ructured connectionist paradigm. This has resulted in t.he building of systems with semantically nameable nodes which perform inferencing by examining t.he pat,t.erns of activation spread.

ALVIS is a reinforcement-based connectionist architecture that learns associative maps in continuous multidimensional environments. The discovered locations of positive and negative reinforcements are recorded in "do be" and "don't be" subnetworks, respectively. The outputs of the subnetworks relevant to the current goal are combined and compared with the current location to produce an error vector. This vector is backpropagated through a motor-perceptual mapping network.

A mathematical model based on the bulbar anatomy and electrophysiology is described. Simulations produce a 35-60 Hz modulated activity coherent across the bulb, mimicing the observed field potentials. The decision states (for the odor information) here can be thought of as stable cycles, rather than point stable states typical of simpler neuro-computing models. Analysis and simulations show that a group of coupled nonlinear oscillators are responsible for the oscillatory activities determined by the odor input, and that the bulb, with appropriate inputs from higher centers, can enhance or suppress the sensitivity to partiCUlar odors. The model provides a framework in which to understand the transform between odor input and the bulbar output to olfactory cortex.

Learning procedures that measure how random perturbations of unit activities correlate with changes in reinforcement are inefficient but simple to implement in hardware. Procedures like back-propagation (Rumelhart, Hinton and Williams, 1986) which compute how changes in activities affect the output error are much more efficient, but require more complex hardware. GEMINI is a hybrid procedure for multilayer networks, which shares many of the implementation advantages of correlational reinforcement procedures but is more efficient. GEMINI injects noise only at the first hidden layer and measures the resultant effect on the output error. A linear network associated with each hidden layer iteratively inverts the matrix which relates the noise to the error change, thereby obtaining the error-derivatives. No back-propagation is involved, thus allowing unknown non-linearities in the system. Two simulations demonstrate the effectiveness of GEMINI.

An improved learning paradigm that offers a significant reduction in computation time during the supervised learning phase is described. It is based on extending the role that the neuron plays in artificial neural systems. Prior work has regarded the neuron as a strictly passive, nonlinear processing element, and the synapse on the other hand as the primary source of information processing and knowledge retention. In this work, the role of the neuron is extended insofar as allowing its parameters to adaptively participate in the learning phase. The temperature of the sigmoid function is an example of such a parameter.

This paper is concerced with the use of error back-propagation in phonetic classification. Our objective is to investigate the basic characteristics of back-propagation, and study how the framework of multi-layer perceptrons can be exploited in phonetic recognition.

Nearly optimal solutions to many combinatorial problems can be found using stochastic simulated annealing. This paper extends the concept of simulated annealing from its original formulation as a Markov process to a new formulation based on mean field theory. Mean field annealing essentially replaces the discrete degrees of freedom in simulated annealing with their average values as computed by the mean field approximation. The net result is that equilibrium at a given temperature is achieved 1-2 orders of magnitude faster than with simulated annealing. A general framework for the mean field annealing algorithm is derived, and its relationship to Hopfield networks is shown. The behavior of MFA is examined both analytically and experimentally for a generic combinatorial optimization problem: graph bipartitioning. This analysis indicates the presence of critical temperatures which could be important in improving the performance of neural networks.

We have fabricated a test chip in 2 micron CMOS technology that embodies these ideas and we report our evaluation of the microchip and our plans for improvements. Knowledge is encoded in the test chip by presenting digital patterns to it that are examples of a desired input-output Boolean mapping. This knowledge is learned and stored entirely on chip in a digitally controlled synapse-like element in the form of connection strengths between neuron-like elements. The only portion of this learning system which is off chip is the VLSI test equipment used to present the patterns. This learning system uses a modified Boltzmann machine algorithm[3] which, if simulated on a serial digital computer, takes enormous amounts of computer time. Our physical implementation is about 100,000 times faster. The test chip, if expanded to a board-level system of thousands of neurons, would be an appropriate architecture for solving artificial intelligence problems whose solutions are hard to specify using a conventional rule-based approach. Examples include speech and pattern recognition and encoding some types of expert knowledge.