Gregory L. Heileman

An ordering algorithm for pattern presentation in fuzzy ARTMAP that tends to improve generalization performance

In this paper we introduce a procedure, based on the max-min clustering method, that identifies a fixed order of training pattern presentation for fuzzy adaptive resonance theory mapping (ARTMAP). This procedure is referred to as the ordering algorithm, and the combination of this procedure with fuzzy ARTMAP is referred to as ordered fuzzy ARTMAP. Experimental results demonstrate that ordered fuzzy ARTMAP exhibits a generalization performance that is better than the average generalization performance of fuzzy ARTMAP, and in certain cases as good as, or better than the best fuzzy ARTMAP generalization performance. We also calculate the number of operations required by the ordering algorithm and compare it to the number of operations required by the training phase of fuzzy ARTMAP. We show that, under mild assumptions, the number of operations required by the ordering algorithm is a fraction of the number of operations required by fuzzy ARTMAP.

Fuzzy ART properties

Abstract This paper presents some important properties of the Fuzzy ART neural network algorithm introduced by Carpenter, Grossberg, and Rosen. The properties described in the paper are distinguished into a number of categories. These include template, access, and reset properties, as well as properties related to the number of list presentations needed for weight stabilization. These properties provide numerous insights as to how Fuzzy ART operates. Furthermore, the effects of the Fuzzy ART parameters a and p on the functionality of the algorithm are clearly illustrated.

Order of search in fuzzy ART and fuzzy ARTMAP: effect of the choice parameter

This paper focuses on two ART architectures, the Fuzzy ART and the Fuzzy ARTMAP. Fuzzy ART is a pattern clustering machine, while Fuzzy ARTMAP is a pattern classification machine. Our study concentrates on the order according to which categories in Fuzzy ART, or the ART(a) model of Fuzzy ARTMAP are chosen. Our work provides a geometrical, and clearer understanding of why, and in what order, these categories are chosen for various ranges of the choice parameter of the Fuzzy ART module. This understanding serves as a powerful tool in developing properties of learning pertaining to these neural network architectures; to strengthen this argument, it is worth mentioning that the order according to which categories are chosen in ART 1 and ARTMAP provided a valuable tool in proving important properties about these architectures. Copyright 1996 Elsevier Science Ltd.

fMRI pattern classification using neuroanatomically constrained boosting.

Pattern classification in functional MRI (fMRI) is a novel methodology to automatically identify differences in distributed neural substrates resulting from cognitive tasks. Reliable pattern classification is challenging due to the high dimensionality of fMRI data, the small number of available data sets, interindividual differences, and dependence on the acquisition methodology. Thus, most previous fMRI classification methods were applied in individual subjects. In this study, we developed a novel approach to improve multiclass classification across groups of subjects, field strengths, and fMRI methods. Spatially normalized activation maps were segmented into functional areas using a neuroanatomical atlas and each map was classified separately using local classifiers. A single multiclass output was applied using a weighted aggregation of the classifiers outputs. An Adaboost technique was applied, modified to find the optimal aggregation of a set of spatially distributed classifiers. This Adaboost combined the region-specific classifiers to achieve improved classification accuracy with respect to conventional techniques. Multiclass classification accuracy was assessed in an fMRI group study with interleaved motor, visual, auditory, and cognitive task design. Data were acquired across 18 subjects at different field strengths (1.5 T, 4 T), with different pulse sequence parameters (voxel size and readout bandwidth). Misclassification rates of the boosted classifier were between 3.5% and 10%, whereas for the single classifier, these were between 15% and 23%, suggesting that the boosted classifier provides a better generalization ability together with better robustness. The high computational speed of boosting classification makes it attractive for real-time fMRI to facilitate online interpretation of dynamically changing activation patterns.

DRM as a layered system

The current landscape for digital rights management(DRM) consists of various ad hoc technologies and platforms that largely focus on copy protection. The fragmented nature of the DRM industry in 2004 is somewhat reminiscent of the telecommunications industry in the late 1980s. At that time various networking technologies were available, and what was needed was a technology that could integrate existing networks and provide various services to users. The OSI layered framework and the TCP/IP communications protocol suite provided a solution to this situation. The OSI model divides the process of digital data communications into layers. Likewise, in this paper we divide the process of DRM into layers in which various services are offered to the users of digital content at each layer. Three blocks of layers have been identified. The upper layers deal with the end-to-end functions of the application, the middle layers deal with rights expression and interpretation, and the lower layers ensure rights enforcement. This paper describes how responsibilities might be distributed among the various layers, and considers where in these layers it would be appropriate to define protocols and standards.

DRM interoperability analysis from the perspective of a layered framework

Interoperability is currently seen as one of the most significant problems facing the digital rights management (DRM) industry. In this paper we consider the problem of interoperability among DRM systems from the perspective of a layered architectural framework. The advantage of looking at the problem from this point of view is that the layered framework provides a certain amount of structure that is very helpful in guiding those working on DRM interoperability issues. Specifically, the layered framework we describe provides a useful design abstraction along architectural lines. One of the advantages of this perspective is that it allows us to consider the level within computing/communication architectures at which certain functionality should be provided, and then to address how the functionality between layers should interact in order to provide specific DRM capabilities. The communications that occur between layers, both within a single system and between two communicating systems, are the places where protocols can be defined and possibly standardized. Thus, they provide focal points for studying and addressing interoperability in DRM systems.

Properties of learning related to pattern diversity in ART1

Abstract In this paper we consider a special class of the ART1 neural network. It is shown that if this network is repeatedly presented with an arbitrary list of binary input patterns, learning self-stabilizes in at most m list presentations, where m corresponds to the number of patterns of distinct size in the input list. Other useful properties of the ART1 network, associated with the learning of an arbitrary list of binary input patterns, are also examined. These properties reveal some of the “good” characteristics of the ART1 network when it is used as a tool for the learning of recognition categories.

Digital rights management architectures

Digital rights management (DRM) is increasingly becoming a necessity for content management and distribution in highly networked environments such as the Internet. However, very few DRM models have been able to achieve commercial success and acceptance among users. This paper analyzes the problems with current DRM environments and proposes an open layered framework for development of DRM systems, where different technologies can interoperate within the framework. Furthermore, interoperability is studied in terms of the proposed layered framework, and problems posed by the current rights expression languages (RELs) are identified. We conclude that a refactoring of current RELs based on a set of design principles is necessary to achieve a reasonable level of DRM interoperability. We emphasize the need for middleware services for DRM, along with their responsibilities and places of operation within the proposed framework. Finally, a specific prototype architecture is introduced that makes use of existing infrastructures in order to implement a DRM environment consistent with the design principles described in this paper.

Multimodal and Multi-Tissue Measures of Connectivity Revealed by Joint Independent Component Analysis

The human brain functions as an efficient system where signals arising from gray matter are transported via white matter tracts to other regions of the brain to facilitate human behavior. However, with a few exceptions, functional and structural neuroimaging data are typically optimized to maximize the quantification of signals arising from a single source. For example, functional magnetic resonance imaging (FMRI) is typically used as an index of gray matter functioning whereas diffusion tensor imaging (DTI) is typically used to determine white matter properties. While it is likely that these signals arising from different tissue sources contain complementary information, the signal processing algorithms necessary for the fusion of neuroimaging data across imaging modalities are still in a nascent stage. In the current paper we present a data-driven method for combining measures of functional connectivity arising from gray matter sources (FMRI resting state data) with different measures of white matter connectivity (DTI). Specifically, a joint independent component analysis (J-ICA) was used to combine these measures of functional connectivity following intensive signal processing and feature extraction within each of the individual modalities. Our results indicate that one of the most predominantly used measures of functional connectivity (activity in the default mode network) is highly dependent on the integrity of white matter connections between the two hemispheres (corpus callosum) and within the cingulate bundles. Importantly, the discovery of this complex relationship of connectivity was entirely facilitated by the signal processing and fusion techniques presented herein and could not have been revealed through separate analyses of both data types as is typically performed in the majority of neuroimaging experiments. We conclude by discussing future applications of this technique to other areas of neuroimaging and examining potential limitations of the methods.

The problem with rights expression languages

In this paper we consider the functionality that a rights expression language (REL) should provide within a digital rights management (DRM) environment. We begin by noting the dearth of applications that make use of RELs, despite the fact that they have now been available since the late 1990s. We posit that one of the main impediments to the use of RELs is the complexity associated with understanding and using them. This results from the fact that the functionality needed to handle a wide variety of possible DRM scenarios is typically built into a REL, and it is often difficult to cleanly partition out only those pieces needed by a particular DRM application. Basing DRM system design on a layered architecture provides one way of achieving a partitioning and points to the need for a simple REL that is exclusively responsible for the expression of rights, while pushing much of the functionality found in current RELs into higher system layers. In order to demonstrate the usefulness of this approach, we provide an example implementation dealing with DRM-based negotiation.