Search

The powers that artificial intelligence (AI) has developed are impressive, with recent success in leveraging human expertise at various stages of model development. AI can attain its full potential only if, as part of its intelligence, it also actively teams with humans to co-create solutions. Combining AI simulation with human intelligence through data convergence can improve decision-making processes and provide a capacity akin to a "teaming intelligence." This research, for the first time, introduces the concepts of Human-AI Convergence (HAC) capabilities for flood evacuation decision-making. The objective of this study was to develop a unique, computationally effective surrogate HAC system for flood evacuation decision-making that integrates the distinctive features of AI with transportation geospatial data, a river hydraulic model, and human data from X (previously Twitter) to visualize flood inundation areas and suggest re-routing. The HAC system is smartly designed to forecast flood stage levels using AI across the US Geological Survey gauging stations and combine the results with Manning's equation results and transportation data, integrated into a web-based Google Earth visualization architecture. The technology has been tested in the Lowcountry of South Carolina, where previous flooding disasters caused considerable damage to the transportation networks and increased traffic on evacuation routes. This state-of-the-art HAC system -- a flood evacuation product -- stands to advance the frontier of human-AI collaborative research in the context of real-time flood emergency management and response. [ABSTRACT FROM AUTHOR]

This document provides an overview of the development of an AI curriculum for gifted students in grades 6-10. The curriculum includes subjects such as introduction to AI, probability, supervised and unsupervised learning, clustering, decision trees, and the Naive Bayes algorithm. The curriculum does not require coding as a prerequisite and emphasizes the importance of mathematical and statistical dimensions in AI education. The document also includes resources related to AI education in Turkey, such as curriculum programs and guidelines developed by the Ministry of Education. Additionally, it discusses the foundations and configuration of Bayesian networks, including calculations of marginal probabilities, joint probabilities, and conditional probabilities. Examples and explanations are provided to aid understanding and application of these concepts. [Extracted from the article]

Pater's (2019) target article proposes that neural networks will provide theories of learning that generative grammar lacks. We argue that his enthusiasm is premature since the biases of neural networks are largely unknown, and he disregards decades of work on machine learning and learnability. Learning biases form a two-way street: all learners have biases, and those biases constrain the space of learnable grammars in mathematically measurable ways. Analytical methods from the related fields of computational learning theory and grammatical inference allow one to study language learning, neural networks, and linguistics at an appropriate level of abstraction. The only way to satisfy our hunger and to make progress on the science of language learning is to confront these core issues directly. [ABSTRACT FROM AUTHOR]

Sparse representations have gained much interest due to the rapid growth of intelligent embedded systems, the need to reduce the time to mine large datasets, and for reducing the footprint of recognition based applications on portable devices. Computational learning theory tells us that the Vapnik-Chervonenkis (VC) dimension of a learning model directly impacts the structural risk and the generalization ability of a learning model. The minimal complexity machine (MCM) was recently proposed as a way to learn a hyperplane classifier by minimizing a tight bound on the VC dimension; results show that it learns very sparse representations that yield test set accuracies that are comparable to the state-of-the-art. The MCM formulation works in the primal itself, both when the classifier is learnt in the input space and when it is learnt implicitly in a higher dimensional feature space. In the latter case, the hyperplane is constructed in the empirical feature space (EFS). In this paper, we examine the hyperplane restricted to the EFS. The EFS is a finite dimensional vector space spanned by the image vectors in a higher dimensional feature space. Since the VC dimension of a linear hyperplane classifier is exactly the number of features used by the classifier, the dimension of the EFS is a direct measure of both the sparsity of the model and the VC dimension. This allows us to formulate optimization problems that focus on learning sparse representations, and yet generalize well. We derive an EFS version of the MCM, that allows us to minimize the model complexity and improve sparsity. We also propose a novel least squares version of the MCM in the EFS. Experimental results demonstrate that the EFS variants yield sparse models with generalization comparable to the state-of-the-art. [ABSTRACT FROM AUTHOR]

Recent advances in image-based object recognition have exploited object proposals to speed up the detection process by reducing the search space. In this paper, we present a novel idea that utilizes true objectness and semantic image filtering (retrieved within the convolutional layers of a Convolutional Neural Network) to propose effective region proposals. Information learned in fully convolutional layers is used to reduce the number of proposals and enhance their localization by producing highly accurate bounding boxes. The greatest benefit of our method is that it can be integrated into any existing approach exploiting edge-based objectness to achieve consistently high recall across various intersection over union thresholds. Experiments on PASCAL VOC 2007 and ImageNet datasets demonstrate that our approach improves two existing state-of-the-art models with significantly high margins and pushes the boundaries of object proposal generation. [ABSTRACT FROM AUTHOR]

Abstract: With the rapid increases in processing speed and memory of low-cost computers, it is not surprising that various advanced computational learning tools such as neural networks have been increasingly used for analyzing or modeling highly nonlinear multivariate engineering problems. These algorithms are useful for analyzing many geotechnical problems, particularly those that lack a precise analytical theory or understanding of the phenomena involved. In situations where measured or numerical data are available, neural networks have been shown to offer great promise for mapping the nonlinear interactions (dependency) between the system’s inputs and outputs. Unlike most computational tools, in neural networks no predefined mathematical relationship between the dependent and independent variables is required. However, neural networks have been criticized for its long training process since the optimal configuration is not known a priori. This paper explores the use of a fairly simple nonparametric regression algorithm known as multivariate adaptive regression splines (MARS) which has the ability to approximate the relationship between the inputs and outputs, and express the relationship mathematically. The main advantages of MARS are its capacity to produce simple, easy-to-interpret models, its ability to estimate the contributions of the input variables, and its computational efficiency. First the MARS algorithm is described. A number of examples are then presented that explore the generalization capabilities and accuracy of this approach in comparison to the back-propagation neural network algorithm. [Copyright &y& Elsevier]

Abstract: Cooperative coevolution divides an optimisation problem into subcomponents and employs evolutionary algorithms for evolving them. Problem decomposition has been a major issue in using cooperative coevolution for neuro-evolution. Efficient problem decomposition methods group interacting variables into the same subcomponents. It is important to find out which problem decomposition methods efficiently group subcomponents and the behaviour of neural network during training in terms of the interaction among the synapses. In this paper, the interdependencies among the synapses are analysed and a problem decomposition method is introduced for feedforward neural networks on pattern classification problems. We show that the neural network training problem is partially separable and that the level of interdependencies changes during the learning process. The results confirm that the proposed problem decomposition method has improved performance compared to its counterparts. [Copyright &y& Elsevier]

Abstract: In this paper, a neural classifier based on the newly developed local coupled feedforward neural network, which may improve the convergence of BP learning significantly, is developed. A binary threshold unit is used as the output node of the classifier. A general error gradient of the output node is defined for the BP training of the classifier. And a hidden node selection scheme is developed for the local coupled feedforward neural network. In addition, we derive a result on the “universal approximation” property of the local coupled feedforward neural network with an arbitrary group of window functions, which can cover the region of training samples. Simulation results show that the general error gradient and the hidden node selection scheme work well. [Copyright &y& Elsevier]

Abstract: In this paper, a system based on artificial neural networks (ANNs) was designed to diagnose different types of fault in a gearbox. An experimental set of data was used to verify the effectiveness and accuracy of the proposed method. The system was optimized by eliminating unimportant features using a feature selection method (UTA method). Consequently, the fault detection system operates faster while the classification error decreases or remains constant in some other cases. This method of feature selection is compared with Genetic Algorithm (GA) results. The findings verify that the results of the UTA method are as accurate as GA, despite its simple algorithm. [Copyright &y& Elsevier]

In this paper, dynamic neural fields (DNFs) are used to develop key features of a cortically-inspired computational module. Under the perspective of designing computational systems that can exhibit the flexibility and genericity of the cortical substrate, using neural field as the competition layer for self-organising modules has to be considered. However, despite the fact that they serve as a biologically-inspired model, applying DNFs to drive self-organisation is not straightforward. In order to address that issue, an original method for evaluating neural field equations is proposed, based on statistical measurements of the field behaviour in some scenarios. Limitations of classical neural field equations are then quantified, and an original field equation is proposed to overcome these difficulties. The performance of the proposed field model is discussed in comparison with some previously considered models, leading to the promotion of the proposed model as a suitable mean for processing competition in cortex-like computation for cognitive systems. [ABSTRACT FROM AUTHOR]

Abstract: Fusing the outputs of an ensemble of diverse predictive models usually boosts overall prediction accuracy. Such fusion is guided by each model''s local performance, i.e., each model''s prediction accuracy in the neighborhood of the probe point. Therefore, for each probe we instantiate a customized fusion mechanism. The fusion mechanism is a meta-model, i.e., a model that operates one level above the object-level models whose predictions we want to fuse. Like these models, such a meta-model is defined by structural and parametric information. In this paper, we focus on the definition of the parametric information for a given structure. For each probe point, we either retrieve or compute the parameters to instantiate the associated meta-model. The retrieval approach is based on a CART-derived segmentation of the probe''s state space, which contains the meta-model parameters. The computation approach is based on a run-time evaluation of each model''s local performance in the neighborhood of the probe. We explore various structures for the meta-model, and for each structure we compare the pre-compiled (retrieval) or run-time (computation) approaches. We demonstrate this fusion methodology in the context of multiple neural network models. However, our methodology is broadly applicable to other predictive modeling approaches. This fusion method is illustrated in the development of highly accurate models for emissions, efficiency, and load prediction in a complex power plant. The locally weighted fusion method boosts the predictive performance by 30–50% over the baseline single model approach for the various prediction targets. Relative to this approach, typical fusion strategies that use averaging or globally weighting schemes only produce a 2–6% performance boost over the same baseline. [Copyright &y& Elsevier]

Contemporary image recognition schemes either rely on single-feature recognition or focus on solving multi-feature recognition using complex computational approaches. Furthermore these approaches tend to be of tightly-coupled nature, thus not readily deployable within computational networks. Distributed Hierarchical Graph Neuron (DHGN) is a distributed single-cycle learning pattern recognition algorithm that can scale from coarse-grained to fine-grained networks and it has comparable accuracy to contemporary image recognition schemes. In this paper, we present an implementation of DHGN that works for multi-feature recognition of images. Our scheme is able to disseminate recognition of each feature within an image to a separate computational subnetwork. Thereby allowing a number of features being analysed simultaneously using a uniform recognition process. We have conducted tests on a collection of greyscale facial images. The results show that our approach produces high recognition accuracy through a simple distributed process. Furthermore, our approach implements single-cycle learning known as collaborative-comparison learning where new patterns are continuously stored using collaborative approach without affecting previously stored patterns. Our proposed scheme demonstrates higher classification accuracy in comparison with Back-Propagation Neural Network for multi-class images. [ABSTRACT FROM AUTHOR]

Abstract: A new efficient technique for estimating probability densities from data through the application of the approximate global maximum likelihood (AGML) approach is proposed. It employs a composition of kernel functions to estimate the correct behavior of parameters involved in the expression of the unknown probability density. Convergence to the optimal solution is guaranteed by a deterministic learning framework when low discrepancy sequences are used to generate the centers of the kernels. Trials on mixture of Gaussians show that the proposed semi-local technique is able to efficiently approximate the maximum likelihood solution even in complex situations where implementations based on standard neural networks require an excessive computational cost. [Copyright &y& Elsevier]

Abstract: The purpose of this paper is to study loss functions in multiclass classification. In classification problems, the decision function is estimated by minimizing an empirical loss function, and then, the output label is predicted by using the estimated decision function. We propose a class of loss functions which is obtained by a deformation of the log-likelihood loss function. There are four main reasons why we focus on the deformed log-likelihood loss function: (1) this is a class of loss functions which has not been deeply investigated so far, (2) in terms of computation, a boosting algorithm with a pseudo-loss is available to minimize the proposed loss function, (3) the proposed loss functions provide a clear correspondence between the decision functions and conditional probabilities of output labels, (4) the proposed loss functions satisfy the statistical consistency of the classification error rate which is a desirable property in classification problems. Based on (3), we show that the deformed log-likelihood loss provides a model of mislabeling which is useful as a statistical model of medical diagnostics. We also propose a robust loss function against outliers in multiclass classification based on our approach. The robust loss function is a natural extension of the existing robust loss function for binary classification. A model of mislabeling and a robust loss function are useful to cope with noisy data. Some numerical studies are presented to show the robustness of the proposed loss function. A mathematical characterization of the deformed log-likelihood loss function is also presented. [Copyright &y& Elsevier]

Artificial neural networks, trained only on sample deals, without presentation of any human knowledge or even rules of the game, are used to estimate the number of tricks to be taken by one pair of bridge players in the so-called double dummy bridge problem (DDBP). Four representations of a deal in the input layer were tested leading to significant differences in achieved results. In order to test networks' abilities to extract knowledge from sample deals, experiments with additional inputs representing estimators of hand's strength used by humans were also performed. The superior network trained solely on sample deals outperformed all other architectures, including those using explicit human knowledge of the game of bridge. Considering the suit contracts, this network, in a sample of 100000 testing deals, output a perfect answer in 53.11% of the cases and only in 3.52% of them was mistaken by more than one trick. The respective figures for notrump contracts were equal to 37.80% and 16.36%. The above results were compared with the ones obtained by 24 professional human bridge players—members of The Polish Bridge Union—on test sets of sizes between 27 and 864 deals per player (depending on player's time availability). In case of suit contracts, the perfect answer was obtained in 53.06% of the testing deals for ten upper-classified players and in 48.66% of them, for the remaining 14 participants of the experiment. For the notruinp contracts, the respective figures were equal to 73.68% and 60.78%. Except for checking the ability of neural networks in solving the DDBP, the other goal of this research was to analyze connection weights in trained networks in a quest for weights' patterns that are explainable by experienced human bridge players. Quite surprisingly, several such patterns were discovered (e.g., preference for groups of honors, drawing special attention to Aces, favoring cards from a trump suit, gradual importance of cards in one suit—from two to the Ace, etc.). Both the numerical figures and weight patterns are stable and repeatable in a sample of neural architectures (differing only by randomly chosen initial weights). In summary, the piece of research described in this paper provides a detailed comparison between various data representations of the DDBP solved by neural networks. On a more general note, this approach can be extended to a certain class of binary classification problems. [ABSTRACT FROM AUTHOR]

Most conventional learning algorithms require both positive and negative training data for achieving accurate classification results. However, the problem of learning classifiers from only positive data arises in many applications where negative data are too costly, difficult to obtain, or not available at all. This paper describes a new machine learning approach, called ILoNDF (Incremental data-driven Learning of Novelty Detector Filter). The approach is inspired by novelty detection theory and its learning method, which typically requires only examples from one class to learn a model. One advantage of ILoNDF is the ability of its generative learning to capture the intrinsic characteristics of the training data by continuously integrating the information relating to the relative frequencies of the features of training data and their co-occurrence dependencies. This makes ILoNDF rather stable and less sensitive to noisy features which may be present in the representation of the positive data. In addition, ILoNDF does not require extensive computational resources since it operates on-line without repeated training, and no parameters need to be tuned. In this study we mainly focus on the robustness of ILoNDF in dealing with high-dimensional noisy data and we investigate the variation of its performance depending on the amount of data available for training. To make our study comparable to previous studies, we investigate four common methods: PCA residuals, Hotelling’s T2 test, an auto-associative neural network, and a one-class version of the SVM classifier (lately a favored method for one-class classification). Experiments are conducted on two real-world text corpora: Reuters and WebKB. Results show that ILoNDF tends to be more robust, is less affected by initial settings, and consistently outperforms the other methods. [ABSTRACT FROM AUTHOR]

In this paper, an up-to-date review is presented of different approaches to the vast field of robot learning. The objective of this paper is to make a robust systematization of knowledge in this area, while focusing on a simple proposed model. Various kinds of learning methods are listed, such as neural networks, empiric learning, genetic algorithms or statistical learning, together with their strong and weak points. Furthermore, aspects are analyzed that make learning subject for mobile robots, while focusing on the most suitable methods for the Powerbot Mobile Robot platform. A model based on the way in which an autonomous mobile robot can use the World Wide Web to determine environment parameters is detailed. Research issues are listed in the last section. [ABSTRACT FROM AUTHOR]

Abstract: We describe a new approach for understanding the difficulty of designing efficient learning algorithms. We prove that the existence of an efficient learning algorithm for a circuit class C in Angluin''s model of exact learning from membership and equivalence queries or in Valiant''s PAC model yields a lower bound against C. More specifically, we prove that any subexponential time, deterministic exact learning algorithm for C (from membership and equivalence queries) implies the existence of a function f in such that . If C is PAC learnable with membership queries under the uniform distribution or exact learnable in randomized polynomial-time, we prove that there exists a function (the exponential time analog of ) such that . For C equal to polynomial-size, depth-two threshold circuits (i.e., neural networks with a polynomial number of hidden nodes), our result shows that efficient learning algorithms for this class would solve one of the most challenging open problems in computational complexity theory: proving the existence of a function in or that cannot be computed by circuits from C. We are not aware of any representation-independent hardness results for learning depth-2, polynomial-size neural networks with respect to the uniform distribution. Our approach uses the framework of the breakthrough result due to Kabanets and Impagliazzo showing that derandomizing yields non-trivial circuit lower bounds. [Copyright &y& Elsevier]

In this brief, a new neural network model called generalized adaptive resonance theory (GART) is introduced. GART is a hybrid model that comprises a modified Gaussian adaptive resonance theory (MGA) and the generalized regression neural network (GRNN). It is an enhanced version of the GRNN, which preserves the online learning properties of adaptive resonance theory (ART). A series of empirical studies to assess the effectiveness of GART in classification, regression, and time series prediction tasks is conducted. The results demonstrate that GART is able to produce good performances as compared with those of other methods, including the online sequential extreme learning machine (OSELM) and sequential learning radial basis function (RBF) neural network models. [ABSTRACT FROM AUTHOR]

Abstract: The radial basis function (RBF) networks are one of the most widely used models for function approximation in the regression problem. In the learning paradigm, the best approximation is recursively or iteratively searched for based on observed data (teacher signals). One encounters difficulties in such a process when two component basis functions become identical, or when the magnitude of one component becomes null. In this case, the number of the components reduces by one, and then the reduced component recovers as the learning process proceeds further, provided such a component is necessary for the best approximation. Strange behaviors, especially the plateau phenomena, have been observed in dynamics of learning when such reduction occurs. There exist singularities in the space of parameters, and the above reduction takes place at the singular regions. This paper focuses on a detailed analysis of the dynamical behaviors of learning near the overlap and elimination singularities in RBF networks, based on the averaged learning equation that is applicable to both on-line and batch mode learning. We analyze the stability on the overlap singularity by solving the eigenvalues of the Hessian explicitly. Based on the stability analysis, we plot the analytical dynamic vector fields near the singularity, which are then compared to those real trajectories obtained by a numeric method. We also confirm the existence of the plateaus in both batch and on-line learning by simulation. [Copyright &y& Elsevier]

Abstract-The recent years have witnessed a surge of interest in semisupervised learning. Numerous methods have been proposed for learning from partially labeled data. In this brief, a novel semisupervised learning approach based on an electrostatic field model is proposed. We treat the labeled data points as point charges, therefore the remaining unlabeled data points are placed in the electrostatic fields generated by these charges. The labels of these unlabeled data points can be regarded as the electric potentials of the electrostatic field at their corresponding places. Moreover, we also develop an efficient way to extend our method for out-of-sample data and analyze theoretically the relationship between our method and the traditional graph-based methods. Finally, the experimental results on both toy and real-world data sets are provided to show the effectiveness of our method. [ABSTRACT FROM AUTHOR]

Artificial immune systems (AIS) are one of the most recent approaches in computational intelligence. This emerging field is sometimes referred to as immunological computation. The development of AIS is fostered by the advances made in understanding the complex mechanisms of the biological immune system. They provide efficient and robust information processing capabilities for solving complex problems. They can learn, adapt previously learned information, and perform pattern recognition/classification in a distributed way. This article discusses AIS especially in the context of misbehavior detection in wireless ad hoc (sensor) networks. [ABSTRACT FROM AUTHOR]

M. M. Botvinick, T. S. Braver, D. M. Barch, C. S. Carter, and J. D. Cohen (2001) implemented their conflict-monitoring hypothesis of cognitive control in a series of computational models. The authors of the current article first demonstrate that M. M. Botvinick et al's (2001) conflict-monitoring Stroop model fails to simulate L. L. Jacoby, D. S. Lindsay, and S. Hessels's (2003) report of an item-specific proportion-congruent (ISPC) effect in the Stroop task. The authors then implement a variant of M. M. Botvinick et al.'s model based on the assumption that control must be able to operate at the item level. This model successfully simulates the ISPC effect. In addition, the model provides an alternative to M. M. Botvinick et al's explanation of the list-level proportion-congruent effect in terms of an ISPC effect. Implications of the present modeling effort are discussed. [ABSTRACT FROM AUTHOR]

From computational learning theory, sample size in machine learning problems indeed affects the learning performance. Since only few samples can be obtained in the early stages of a system and fewer exemplars usually lead to a low learning accuracy, this research compares different machine learning methods through their classification accuracies to improve small-data-set learning. Techniques used in this paper include the mega-trend diffusion technique, a backpropagation neural network, a support vector machine, and decision trees to explore the machine learning issue with two real medical data sets concerning cancer. The result of the experiment shows that the mega-trend diffusion technique and backpropagation approaches are effective methods of small-data-set learning. [ABSTRACT FROM AUTHOR]

A new approach based on a synergetic combination of statistical/machine learning and deterministic modeling within atmospheric models is presented. The approach uses neural networks as a statistical or machine learning technique for an accurate and fast emulation or statistical approximation of model physics parameterizations. It is applied to development of an accurate and fast approximation of an atmospheric longwave radiation parameterization for the NCAR Community Atmospheric Model, which is the most time consuming component of model physics. The developed neural network emulation is two orders of magnitude, 50–80 times, faster than the original parameterization. A comparison of the parallel 10-yr climate simulations performed with the original parameterization and its neural network emulations confirmed that these simulations produce almost identical results. The obtained results show the conceptual and practical possibility of an efficient synergetic combination of deterministic and statistical learning components within an atmospheric climate or forecast model. A developmental framework and practical validation criteria for neural network emulations of model physics components are outlined. [ABSTRACT FROM AUTHOR]

The question of whether it is possible to load deep neural network architectures efficiently is examined by considering the class of pyramidal architectures. This class allows only a low interaction of the nodes. Still, the loading problem is found to be NP-complete. This provides evidence that depth alone is a factor accounting for loading hardness. [ABSTRACT FROM AUTHOR]

We show how Bayesian neural networks can be used for time-series analysis. We consider a block-based model building strategy to model linear and nonlinear features within the time series: a linear combination of a linear autoregression term and a feedforward neural network (FFNN) with an unknown number of hidden nodes. To allow for simpler models, we also consider these terms separately as competing models to select from. Model identifiability problems arise when FFNN sigmoidal activation functions exhibit almost linear behavior or when there are almost duplicate or irrelevant neural network nodes. New reversible-jump moves are proposed to facilitate model selection, mitigating model identifiability problems. We illustrate this methodology analyzing several time-series data examples. [ABSTRACT FROM AUTHOR]

We study networks of spiking neurons that use the timing of pulses to encode information. Nonlinear interactions model the spatial groupings of synapses on the neural dendrites and describe the computations performed at local branches. Within a theoretical framework of learning we analyze the question of how many training examples these networks must receive to be able to generalize well. Bounds for this sample complexity of learning can be obtained in terms of a combinatorial parameter known as the pseudodimension. This dimension characterizes the computational richness of a neural network and is given in terms of the number of network parameters. Two types of feedforward architectures are considered: constant-depth networks and networks of unconstrained depth. We derive asymptotically tight bounds for each of these network types. Constant depth networks are shown to have an almost linear pseudodimension, whereas the pseudodimension of general networks is quadratic. Networks of spiking neurons that use temporal coding are becoming increasingly more important in practical tasks such as computer vision, speech recognition, and motor control. The question of how well these networks generalize from a given set of training examples is a central issue for their successful application as adaptive systems. The results show that, although coding and computation in these networks is quite different and in many cases more powerful, their generalization capabilities are at least as good as those of traditional neural network models. [ABSTRACT FROM AUTHOR]

Recurrent networks can generate spatio-temporal neural sequences of very large cycles, having an apparent random behavior. Nonetheless a proximity measure between these sequences may be defined through comparison of the synaptic weight. matrices that generate them. Following the dynamic neural filter (DNF) formalism we demonstrate this concept by comparing teacher and student recurrent networks of binary neurons. We show that large sequences, providing a training set well exceeding the Cover limit, allow for good determination of the synaptic matrices. Alternatively, assuming the matrices to be known, very fast determination of the biases can be achieved. Thus, a spatio-temporal sequence may be regarded as spatio-temporal encoding of the bias vector. We introduce a linear support vector machine (SVM) variant of the DNF in order to specify an optimal weight matrix. This approach allows us to deal with noise. Spatio-temporal sequences generated by different DNFs with the same number of neurons may be compared by calculating correlations of the synaptic matrices of the reconstructed DNFs. Other types of spatiotemporal sequences need the introduction of hidden neurons, and/or the use of a kernel variant of the SYM approach. The latter is being defined as a recurrent support vector network (RSVN). [ABSTRACT FROM AUTHOR]

We discuss the participatory learning paradigm and emphasize the all?inclusive role that what is already believed plays in every aspect of the learning process. One manifestation of this is the appearance of an acceptance mechanism. This mechanism determines which data (experiences) will be used for learning based on its compatibility with what we already believe. A prototypical example of a participatory learning system is introduced. Methods for formulating the required compatibility function are discussed. We introduce the concept of a critic that monitors the performance of the system and gives us some independence from our current beliefs. Finally, we consider a participatory learning system which does not require that the attribute being learned take numeric values. We emphasize that understanding this paradigm can be useful both to help build a digital learning algorithm as well as to provide an instrument for enhanced interaction with others. [ABSTRACT FROM AUTHOR]

Discusses the use of simple synchrony networks (SSN) for learning to parse English sentences drawn from a corpus of naturally occurring text. Incremental representation of constituent structures; Retention of a bounded number of constituents during parsing using a short-term memory mechanism; Improvement of the O(sup n) speed of SSN architecture.

Examines the information theoretical learnability of folding networks. Improvement of VC, pseudo- and fat shattering dimension with various activation networks; Deviation of the empirical from the real error of a learning algorithm.

The guest editor discusses some recent advances in machine learning and their applications to exciting new problem areas. [ABSTRACT FROM AUTHOR]

Points out that the performance of radio transmitter identification systems for low signal-to-noise ratio (SNR) transients can be improved by utilizing the transient SNR information in the classifier training stage. Estimation of the SNR of the unknown transient, which determines the amount of noise to be added into the neural network training set; Verification of performance improvement observed in simulations with experimental data.