Learning the dynamics of embedded clauses

Recent work by Siegelmann has shown that the computational power of recurrent neural networks matches that of Turing Machines. One important implication is that complex language classes (infinite languages with embedded clauses) can be represented in neural networks. Proofs are based on a fractal encoding of states to simulate the memory and operations of stacks. In the present work, it is shown that similar stack-like dynamics can be learned in recurrent neural networks from simple sequence prediction tasks. Two main types of network solutions are found and described qualitatively as dynamical systems: damped oscillation and entangled spiraling around fixed points. The potential and limitations of each solution type are established in terms of generalization on two different context-free languages. Both solution types constitute novel stack implementations - generally in line with Siegelmann's theoretical work - which supply insights into how embedded structures of languages can be handled in analog hardware.

Teleonomic entropy: measuring the phase-space of end-directed systems

We introduce a novel way of measuring the entropy of a set of values undergoing changes. Such a measure becomes useful when analyzing the temporal development of an algorithm designed to numerically update a collection of values such as artificial neural network weights undergoing adjustments during learning. We measure the entropy as a function of the phase-space of the values, i.e. their magnitude and velocity of change, using a method based on the abstract measure of entropy introduced by the philosopher Rudolf Carnap. By constructing a time-dynamic two-dimensional Voronoi diagram using Voronoi cell generators with coordinates of value- and value-velocity (change of magnitude), the entropy becomes a function of the cell areas. We term this measure teleonomic entropy since it can be used to describe changes in any end-directed (teleonomic) system. The usefulness of the method is illustrated when comparing the different approaches of two search algorithms, a learning artificial neural network and a population of discovering agents. (C) 2004 Elsevier Inc. All rights reserved.

The Politics of Permaculture: Towards an Ecologically Situated Approach to Resource Policy and Governance

The contributions to environmental politics of Torgerson, Oelschlager, Dryzek, Harrè and others, converge in their respective acknowledgements that the shift towards an ‘ecologically situated’ approach to environmental policy, including resource governance, will require the emergence and consolidation of a new lingua franca of environmental discourse. In this paper, I suggest that an extrapolation of permaculture ethics may provide a gambit through which such a discourse may be assembled and organised. I examine six key signifying elements derived from Orr and Capra’s approach to ecological literacy – network, nested system, flow, cycle, development and dynamic balance – and explore the implications that these might have for resource governance and policy, including the (re)framing of assessment indicators, energy auditing, resource management and integrated planning and development.

A power system control scheme based on security visualisation in parameter space

Power system real time security assessment is one of the fundamental modules of the electricity markets. Typically, when a contingency occurs, it is required that security assessment and enhancement module shall be ready for action within about 20 minutes’ time to meet the real time requirement. The recent California black out again highlighted the importance of system security. This paper proposed an approach for power system security assessment and enhancement based on the information provided from the pre-defined system parameter space. The proposed scheme opens up an efficient way for real time security assessment and enhancement in a competitive electricity market for single contingency case

Cadence, war and security

An increased incidence of attack has been identified as a major characteristic of the new threat posed by terrorist groups such as al Qaeda. This article considers what such a change means for Western national security systems by examining bow different parts of the system change over time. It becomes evident that Western national security systems are structured on an assumption of comparatively slow state-based threats. In contrast, terrorist franchises operate at a faster pace, are more 'lightweight' and can adapt within the operational and capability cycles of Western governments. Neither network-centric warfare nor an improved assessment of the threat, called for by some, offers a panacea in this regard. Rather, it is clear that not only do Western governments need to adjust their operational and capability cycles, but that they also need a greater diversity of responses to increase overall national security resilience and offer more tools for policy-makers.

Tracking the states of a nonlinear system in the weight-space of a feed-forward neural network

Nonlinear, non-stationary signals are commonly found in a variety of disciplines such as biology, medicine, geology and financial modeling. The complexity (e.g. nonlinearity and non-stationarity) of such signals and their low signal to noise ratios often make it a challenging task to use them in critical applications. In this paper we propose a new neural network based technique to address those problems. We show that a feed forward, multi-layered neural network can conveniently capture the states of a nonlinear system in its connection weight-space, after a process of supervised training. The performance of the proposed method is investigated via computer simulations.

RSA-Grid: A grid computing based framework for power system reliability and security analysis

Grid computing is an emerging technology for providing the high performance computing capability and collaboration mechanism for solving the collaborated and complex problems while using the existing resources. In this paper, a grid computing based framework is proposed for the probabilistic based power system reliability and security analysis. The suggested name of this computing grid is Reliability and Security Grid (RSA-Grid). Then the architecture of this grid is presented. A prototype system has been built for further development of grid-based services for power systems reliability and security assessment based on probabilistic techniques, which require high performance computing and large amount of memory. Preliminary results based on prototype of this grid show that RSA-Grid can provide the comprehensive assessment results for real power systems efficiently and economically.

Comparison of BR and QR Eigenvalue Algorithms for Power System Small Signal Stability Analysis

The BR algorithm is a novel and efficient method to find all eigenvalues of upper Hessenberg matrices and has never been applied to eigenanalysis for power system small signal stability. This paper analyzes differences between the BR and the QR algorithms with performance comparison in terms of CPU time based on stopping criteria and storage requirement. The BR algorithm utilizes accelerating strategies to improve its performance when computing eigenvalues of narrowly banded, nearly tridiagonal upper Hessenberg matrices. These strategies significantly reduce the computation time at a reasonable level of precision. Compared with the QR algorithm, the BR algorithm requires fewer iteration steps and less storage space without depriving of appropriate precision in solving eigenvalue problems of large-scale power systems. Numerical examples demonstrate the efficiency of the BR algorithm in pursuing eigenanalysis tasks of 39-, 68-, 115-, 300-, and 600-bus systems. Experiment results suggest that the BR algorithm is a more efficient algorithm for large-scale power system small signal stability eigenanalysis.

Analysis of small signal stability margins using genetic optimization

Power system small signal stability analysis aims to explore different small signal stability conditions and controls, namely: (1) exploring the power system security domains and boundaries in the space of power system parameters of interest, including load flow feasibility, saddle node and Hopf bifurcation ones; (2) finding the maximum and minimum damping conditions; and (3) determining control actions to provide and increase small signal stability. These problems are presented in this paper as different modifications of a general optimization to a minimum/maximum, depending on the initial guesses of variables and numerical methods used. In the considered problems, all the extreme points are of interest. Additionally, there are difficulties with finding the derivatives of the objective functions with respect to parameters. Numerical computations of derivatives in traditional optimization procedures are time consuming. In this paper, we propose a new black-box genetic optimization technique for comprehensive small signal stability analysis, which can effectively cope with highly nonlinear objective functions with multiple minima and maxima, and derivatives that can not be expressed analytically. The optimization result can then be used to provide such important information such as system optimal control decision making, assessment of the maximum network's transmission capacity, etc. (C) 1998 Elsevier Science S.A. All rights reserved.

Effects of beta-escin and saponin on the transverse-tubular system and sarcoplasmic reticulum membranes of rat and toad skeletal muscle

Mechanically skinned skeletal muscle fibres from rat and toad were exposed to the permeabilizing agents beta-escin and saponin. The effects of these agents on the sealed transverse tubular system (t-system) and sarcoplasmic reticulum (SR) were examined by looking at changes in the magnitude of the force responses to t-system depolarization, the time course of the fluorescence of fura-2 trapped in the sealed t-system, and changes in the magnitude of caffeine-induced contractures following SR loading with Ca2+ under defined conditions. In the presence of 2 mu g ml(-1) beta-escin and saponin, the response to t-system depolarization was not completely abolished, decreasing to a plateau, and a large proportion of fura-2 remained in the sealed t-system. At 10 mu g ml(-1), both agents abolished the ability of both rat and toad preparations to respond to t-system depolarization after 3 min of exposure, but a significant amount of fura-2 remained in sealed t-tubules even after exposure to 100 mu g ml(-1) beta-escin and saponin for 10 min. beta-Escin took longer than saponin to reduce the t-system depolarizations and fura-2 content of the sealed t-system to a similar level. The ability of the SR to load Ca2+ was reduced to a lower level after treatment with beta-escin than saponin. This direct effect on the SR occurred at much lower concentrations for rat (2 mu g ml(-1) beta-escin and 10 mu g ml(-1) saponin) than toad (10 mu g ml(-1) beta-escin and 150 mu g ml(-1) saponin). The reverse order in sensitivities to beta-escin and saponin of t-system and SR membranes indicates that the mechanisms of action of beta-escin and saponin are different in the two types of membrane. In conclusion, this study shows that: (1) beta-escin has a milder action on the surface membrane than saponin; (2) beta-escin is a more potent modifier of SR function; (3) simple permeabilization of membranes is not sufficient to explain the effects of beta-escin and saponin on muscle membranes; and (4) the t-system network within muscle fibres is not a homogeneous compartment.

Attentional modulation of neural responses to action observation: Implications for models of the human 'mirror' system

In primates, the observation of meaningful, goaldirected actions engages a network of cortical areas located within the premotor and inferior parietal lobules. Current models suggest that activity within these regions arises relatively automatically during passive action observation without the need for topdown control. Here we used functional magnetic resonance imaging to determine whether cortical activit)' associated with action observation is modulated by the strategic allocation of selective attention. Normal observers viewed movie clips of reach-to-grasp actions while performing an easy or difficult visual discrimination at the fovea. A wholebrain analysis was performed to determine the effects of attentional load on neural responses to observed hand actions. Our results suggest that cortical areas involved in action observation are significantiy modulated by attentional load. These findings have important implications for recent attempts to link the human action-observation system to response properties of "mirror neurons" in monkeys.