Abstract and reduced-context representations in fault-finding training

The ability to transfer problem-solving expertise to new situations is regarded as an important practical skill but it is known to be sensitive to contextual differences. Employing abstract and reduced-context training tasks has been proposed as a method of fostering transferable skill but the evidence supporting this approach is limited. The research reported here aimed at assessing the effectiveness of training with diagnostic tasks of varying degrees of contextual detail. The criterion test used to measure training effectiveness was a context-rich diagnostic task that involved locating faults in a representation of a water supply and drainage system (WSDS). In the first study reported (baseline study), subjects were trained and tested on the WSDS task. Various training interventions were employed and a set of diagnostic heuristics proved to be the most effective. Two pilot studies were then carried out to determine whether training with abstract diagnostic tasks could produce transfer to the criterion test, the WSDS. Little evidence of transfer was found; therefore, it was decided to incorporate some context into training but not so much as to obscure important, conceptual task features. This was achieved by constructing reduced-context representations (RCO) of the WSDS at more than one level of abstraction (main study). Training with those representations took less time than training with the WSDS but transfer to the criterion test (the WSDS) was equally efficient. During training with the RCO representations, a technique of refusing costly test information was employed which is thought to have helped subjects discover efficient search strategies and consistently transfer them to the criterion test. This was contrasted with the verbal diagnostic heuristics of the baseline study, which had proved to be effective, but which were not always transferred consistently to the criterion test. This raises the question as to whether diagnostic strategies are necessarily best conveyed solely in verbal form, if the intention is to train transferable skill.     

Abstract representations for interactive visualization of virtual 3D city models

Virtual 3D city models increasingly cover whole city areas; hence, the perception of complex urban structures becomes increasingly difficult. Using abstract visualization, complexity of these models can be hidden where its visibility is unnecessary, while important features are maintained and highlighted for better comprehension and communication. We present a technique to automatically generalize a given virtual 3D city model consisting of building models, an infrastructure network and optional land coverage data; this technique creates several representations of increasing levels of abstraction. Using the infrastructure network, our technique groups building models and replaces them with cell blocks, while preserving local landmarks. By computing a landmark hierarchy, we reduce the set of initial landmarks in a spatially balanced manner for use in higher levels of abstraction. In four application examples, we demonstrate smooth visualization of transitions between precomputed representations; dynamic landmark highlighting according to virtual camera distance; an implementation of a cognitively enhanced route representation, and generalization lenses to combine precomputed representations in focus + context visualization.     

Abstract error groups via Jones unitary braid group representations at q = i

In this paper, we classify a type of abstract groups by the central products of dihedral groups and quaternion groups. We recognize them as abstract error groups which are often not isomorphic to the Pauli groups in the literature. We show the corresponding nice error bases equivalent to the Pauli error bases modulo phase factors. The extension of these abstract groups by the symmetric group are finite images of the Jones unitary representations (or modulo a phase factor) of the braid group at q = i or r = 4. We hope this work can finally lead to new families of quantum error correction codes via the representation theory of the braid group.      

Accelerating the training of feedforward neural networks usinggeneralized Hebbian rules for initializing the internal representations

This paper presents an unsupervised learning scheme for initializing the internal representations of feedforward neural networks, which accelerates the convergence of supervised learning algorithms. It is proposed in this paper that the initial set of internal representations can be formed through a bottom-up unsupervised learning process applied before the top-down supervised training algorithm. The synaptic weights that connect the input of the network with the hidden units can be determined through linear or nonlinear variations of a generalized Hebbian learning rule, known as Oja's rule. Various generalized Hebbian rules were experimentally tested and evaluated in terms of their effect on the convergence of the supervised training process. Several experiments indicated that the use of the proposed initialization of the internal representations significantly improves the convergence of gradient-descent-based algorithms used to perform nontrivial training tasks. The improvement of the convergence becomes significant as the size and complexity of the training task increase.     

本期摘要

新能源汽车节能减排技术研究进展（1）——发动机系统 综述了电动力汽车（EV）、混合动力汽车（HEV）和燃料电池汽车（FCV）节能减排技术研究进展,主要从以下三方面论述：发动机系统、能量管理系统、传感器系统,并指出了存在问题和发展方向。本文主要介绍了EV、HEV和FCV的发动机系统,包括关键技术、电机设计和驱动技术、控制技术。     

Distrbution and Abstract Types in Emerald

Emerald is an object-based language for programming distributed subsystems and applications. Its novel features include 1) a single object model that is used both for programming in the small and in the large, 2) support for abstract types, and 3) an explicit notion of object location and mobility. This paper outlines the goals of Em-erald, relates Emerald to previous work, and describes its type system and distribution support. We are currently constructing a prototype implementation of Emerald.     

Curve representations in multiple resolutions

Both a linear and a spatial representation scheme are derived from binary curve relations. The linear scheme describes a curve by a simple chain code, called Ruli. The spatial approach works with the curve relations in a pyramid. The procedures reducing the resolution of both representations possess the length reduction property.     

Measuring distribution in distributed representations

The concept of distribution is often encountered in neural network architectures without any formal quantification. A method of quantifying the amount of distribution present in the hidden layer representations of a feed-forward network with binary inputs is described.     

Redundant representations in evolutionary computation

This paper discusses how the use of redundant representations influences the performance of genetic and evolutionary algorithms. Representations are redundant if the number of genotypes exceeds the number of phenotypes. A distinction is made between synonymously and non-synonymously redundant representations. Representations are synonymously redundant if the genotypes that represent the same phenotype are very similar to each other. Non-synonymously redundant representations do not allow genetic operators to work properly and result in a lower performance of evolutionary search. When using synonymously redundant representations, the performance of selectorecombinative genetic algorithms (GAs) depends on the modification of the initial supply. We have developed theoretical models for synonymously redundant representations that show the necessary population size to solve a problem and the number of generations goes with O(2(kr)/r), where kr is the order of redundancy and r is the number of genotypic building blocks (BB) that represent the optimal phenotypic BB. As a result, uniformly redundant representations do not change the behavior of GAs. Only by increasing r, which means overrepresenting the optimal solution, does GA performance increase. Therefore, non-uniformly redundant representations can only be used advantageously if a-priori information exists regarding the optimal solution. The validity of the proposed theoretical concepts is illustrated for the binary trivial voting mapping and the real-valued link-biased encoding. Our empirical investigations show that the developed population sizing and time to convergence models allow an accurate prediction of the empirical results.     

Genotype representations in grammatical evolution

Grammatical evolution (GE) is a form of grammar-based genetic programming. A particular feature of GE is that it adopts a distinction between the genotype and phenotype similar to that which exists in nature by using a grammar to map between the genotype and phenotype. Two variants of genotype representation are found in the literature, namely, binary and integer forms. For the first time we analyse and compare these two representations to determine if one has a performance advantage over the other. As such this study seeks to extend our understanding of GE by examining the impact of different genotypic representations in order to determine whether certain representations, and associated diversity-generation operators, improve GE’s efficiency and effectiveness. Four mutation operators using two different representations, binary and gray code representation, are investigated. The differing combinations of representation and mutation operator are tested on three benchmark problems. The results provide support for the use of an integer-based genotypic representation as the alternative representations do not exhibit better performance, and the integer representation provides a statistically significant advantage on one of the three benchmarks. In addition, a novel wrapping operator for the binary and gray code representations is examined, and it is found that across the three problems examined there is no general trend to recommend the adoption of an alternative wrapping operator. The results also back up earlier findings which support the adoption of wrapping.     

Propositional and terminological knowledge representations

Abstract

Formalisms based on a propositional representation demonstrate a high representational competence because of their flexibility, properties, and their great expression power, even if they often fail to represent the meaning of the nodes that make up the propositional networks. The presence of a network of definitions would make the representation richer, and would augment the expressiveness and the inferential capabilities of the whole system: the twofold architecture of the hybrid systems seems to be suitable to satisfy these prerequisites. Nevertheless, the hybrid systems fail to clearly define the intensional and extensional levels of knowledge. This paper proposes a different distribution of tasks between the two parts of a hybrid system on the ground of a precise distinction among intensional, mental extensional, and extensional aspects of knowledge. Moreover, it suggests employing a propositional approach also for arranging the abstract definitions of the terminological box. These principles have been tested in KRAM (Knowledge Representation for Agency Modelling), a hybrid system used as a component both of a cooperative interface and of a system able to model actions and minds of cognitive agents.