Daniel Spiegel

Issues in the instantiation of template classes

Teaching students to incorporate template classes into their C++ projects is an important concept in object-oriented programming. The most efficient implementation method for template classes is dependent on several factors. Two significant factors are different integrated development environments with differing requirements, and incongruous philosophies among instructors. Herein, several template class instantiation methods under Gnu compilers will be discussed, along with their pros and cons.

Sparse data in the evolutionary generation of fuzzy models

Fuzzy rule bases have proven to be an effective tool for modeling complex systems and approximating functions. Two approaches, global and local rule generation, have been identified for the evolutionary generation of fuzzy models. In the global approach, the standard method of employing evolutionary techniques in fuzzy rule base generation, the fitness evaluation of a rule base aggregates the performance of the model over the entire space into a single value. A local fitness assessment utilizes the limited scope of a fuzzy rule to evaluate the performance in regions of the input space. Regardless of the method employed, the ability to construct models is inhibited when training data are sparse. In this research, a multi-criteria fitness function is introduced to incorporate a bias towards smoothness into the evolutionary selection process. Several multi-criteria fitness functions, which differ in the extent of the assessment smoothness and the range of its application, are examined. A set of experiments has been performed to demonstrate the effectiveness of the multi-criteria strategies for the evolutionary generation of fuzzy models with sparse data.

Employing locality in the evolutionary generation of fuzzy rule bases

Fuzzy rule bases have proven to be an effective tool for modeling complex systems and approximating functions. The generation of a fuzzy rule base has generally been accomplished by a heuristic analysis of the relationships of the underlying system or by algorithmic rule generation from training data. Automatic rule generation has utilized clustering algorithms, proximity analysis, and evolutionary techniques to identify approximate relationships between the input and the output. In this research, two general approaches for the evolutionary generation of fuzzy rules are identified and compared: global and local rule generation. Global rule production, which is the standard method of employing evolutionary techniques in fuzzy rule base generation, considers an entire rule base as an element of population. The fitness evaluation of a rule base aggregates the performance of the model over the entire space into a single value. The local approach utilizes the limited scope of a fuzzy rule to evaluate performance in regions of the input space. The local generation of rule bases employs an independent evolutionary search in each region and combines the local results to produce a global model. An experimental suite has been developed to compare the effectiveness of the two strategies for the evolutionary generation of fuzzy models.

Evolutionary strategies for fuzzy models: local vs global construction

This paper presents a framework for studying the effectiveness of evolutionary strategies for generating fuzzy rule bases from training data. The fitness measure needed for selection is obtained by a comparison of the training data with the function approximation defined by a fuzzy rule base. The properties of employing both global and local fitness measures are examined. Rule base completion is obtained by incorporating a global evaluation of the smoothness of the transitions between local regions into the selection process.

Macrophages from lupus-prone MRL mice have a conditional signaling abnormality that leads to dysregulated expression of numerous genes.

Macrophages (mϕ) from pre-diseased mice of the major murine inbred models of spontaneous autoimmunity (AI), including multiple lupus-prone strains and the type I diabetes-prone NOD (non-obese diabetic) strain, have identical apoptotic target-dependent abnormalities. This characteristic feature of mϕ from AI-prone mice suggests that abnormal signaling events induced within mϕ following their interaction with apoptotic targets may predispose to AI. Such signaling abnormalities would affect predominantly the processing and presentation of self-antigen (i.e., derived from apoptotic targets), while sparing the processing and presentation of foreign antigen (i.e., derived from non-apoptotic sources). Here, we used DNA microarrays to test the hypothesis that mϕ from AI-prone mice (MRL/MpJ [MRL/+] or MRL/MpJ-Tnfrsf6lpr [MRL/lpr]) differentially express multiple genes in comparison to non-AI mϕ (BALB/c), but do so in a largely apoptotic cell-dependent manner. Mϕ were stimulated with lipopolysaccharide, a potent innate stimulus, in the presence or absence of serum (an experimental surrogate for apoptotic targets). In accord with our hypothesis, the number of genes differentially expressed by MRL mϕ was significantly increased in the presence vs. the absence of serum, the apoptotic target surrogate (n = 401 vs. n = 201). Notably, for genes differentially expressed by MRL mϕ in the presence of serum, serum-free culture normalized their expression to a level statistically indistinguishable from that by non-AI mϕ. Comparisons of mϕ from AI-prone NOD and non-AI C57BL/6 mice corroborated these findings. Together, these data support the hypothesis that mϕ from MRL and other AI-prone mice are characterized by a conditional abnormality elicited by serum lipids or apoptotic targets.

Tuning membership functions in local evolutionary learning of fuzzy rule bases

The local evolutionary generation of fuzzy rule bases employs independent searches in local regions throughout the input space and combines the local results to produce a global model. The paper presents a rule base tuning strategy that is compatible with the local evolutionary generation of fuzzy rule bases. Rule base tuning is accomplished by modifying the decomposition of the input domain based on the distribution and values of the training data. A local tuning algorithm must maintain a correspondence between competing rules in the population. An experimental suite has been developed to exhibit the potential for model optimization using rule base tuning. of particular interest is the ability of rule base tuning to compensate for the effects of sparse data.

Compensating for sparse data in evolutionary generation of fuzzy models

Evolutionary techniques have proven to be a successful strategy for generating fuzzy rule bases from training data. The locality of fuzzy decompositions permits a local evolutionary strategy consisting of an independent evolutionary generation of each rule. The fitness of a rule is determined by the training data within a neighborhood called the region of inclusion of the rule. When the amount of training data is limited, some local regions may not contain training data. This research examines the feasibility of adding a secondary criterion to the fitness measure to compensate for sparse data. A smoothness measure is computed for each region by comparing the approximating function within the region with those in adjacent regions. Several methods of incorporating the smoothness measure into the fitness evaluation are compared.

Generation of fuzzy models via evolutionary strategies

This paper presents a framework for studying the effectiveness of evolutionary strategies for generating fuzzy rule bases and function approximations from training data. To facilitate the evolutionary operations that modify the elements of the population, a fuzzy rule base is represented as a real-valued matrix. A comparison of the training data with the function approximation associated with a fuzzy rule base provides a measure of agreement of the rule base with the training data. The analysis of training data provides the ability to generate both global and local fitness assessments. The effectiveness of incorporating local information into the evolutionary search is demonstrated by comparing the generation of rule consequences using the global and local strategies.

Evolutionary strategies for generation of fuzzy rule bases: a local approach

Fuzzy rule bases provide a tool for modeling complex systems and approximating functions. Originally, heuristic analysis by experts was used to produce fuzzy models. Recently, algorithms have been developed to produce models from training data. In this research, two general approaches for evolutionary generation of fuzzy rules are identified and compared: global and local reproduction. Global reproduction, which is the standard approach, considers an entire rule base in performing fitness evaluation and regeneration. The local approach considers a series of independent evolutionary selections and produces a model by combining the localized results. An experimental suite has been developed to compare the effectiveness of the approaches in generating models. The parameters considered include the size office training set and the number of rules.