A principle for learning egocentric-allocentric transformation

Numerous single-unit recording studies have found mammalian hippocampal neurons that fire selectively for the animal's location in space, independent of its orientation. The population of such neurons, commonly known as place cells, is thought to maintain an allocentric, or orientation-independent, internal representation of the animal's location in space, as well as mediating long-term storage of spatial memories. The fact that spatial information from the environment must reach the brain via sensory receptors in an inherently egocentric, or viewpoint-dependent, fashion leads to the question of how the brain learns to transform egocentric sensory representations into allocentric ones for long-term memory storage. Additionally, if these long-term memory representations of space are to be useful in guiding motor behavior, then the reverse transformation, from allocentric to egocentric coordinates, must also be learned. We propose that orientation-invariant representations can be learned by neural circuits that follow two learning principles: minimization of reconstruction error and maximization of representational temporal inertia. Two different neural network models are presented that adhere to these learning principles, the first by direct optimization through gradient descent and the second using a more biologically realistic circuit based on the restricted Boltzmann machine (Hinton, 2002; Smolensky, 1986). Both models lead to orientation-invariant representations, with the latter demonstrating place-cell-like responses when trained on a linear track environment.     

Unsupervised neural network learning procedures for feature extraction and classification

In this article, we review unsupervised neural network learning procedures which can be applied to the task of preprocessing raw data to extract useful features for subsequent classification. The learning algorithms reviewed here are grouped into three sections: information-preserving methods, density estimation methods, and feature extraction methods. Each of these major sections concludes with a discussion of successful applications of the methods to real-world problems.The first author is supported by research grants from the James S. McDonnell Foundation (grant #93–95) and the Natural Sciences and Engineering Research Council of Canada. For part of this work, the second author was supported by a Temporary Lectureship from the Academic Initiative of the University of London, and by a grant (GR/J38987) from the Science and Engineering Research Council (SERC) of the UK.     

Modeling mental navigation in scenes with multiple objects

Various lines of evidence indicate that animals process spatial information regarding object locations differently from spatial information regarding environmental boundaries or landmarks. Following Wang and Spelke's (2002) observation that spatial updating of egocentric representations appears to lie at the heart of many navigational tasks in many species, including humans, we postulate a neural circuit that can support this computation in parietal cortex, assuming that egocentric representations of multiple objects can be maintained in prefrontal cortex in spatial working memory (not simulated here). Our method is a generalization of an earlier model by Droulez and Berthoz (1991), with extensions to support observer rotation. We can thereby simulate perspective transformation of working memory representations of object coordinates based on an egomotion signal presumed to be generated via mental navigation. This biologically plausible transformation would allow a subject to recall the locations of previously viewed objects from novel viewpoints reached via imagined, discontinuous, or disoriented displacement. Finally, we discuss how this model can account for a wide range of experimental findings regarding memory for object locations, and we present several predictions made by the model.     

A model for the evaluation of environmental impact indicators for a sustainable maritime transportation systems

Maritime shipping is considered the most efficient, low-cost means for transporting large quantities of freight over significant distances. However, this process also causes negative environmental and societal impacts. Therefore, environmental sustainability is a pressing issue for maritime shipping management, given the interest in addressing important issues that affect the safety, security, and air and water quality as part of the efficient movement of freight throughout the coasts and waterways and associated port facilities worldwide. In-depth studies of maritime transportation systems (MTS) can be used to identify key environmental impact indicators within the transportation system. This paper develops a tool for decision making in complex environments; this tool will quantify and rank preferred environmental impact indicators within a MTS. Such a model will help decision-makers to achieve the goals of improved environmental sustainability. The model will also provide environmental policy-makers in the shipping industry with an analytical tool that can evaluate tradeoffs within the system and identify possible alternatives to mitigate detrimental effects on the environment.     

Nowcasting model of low wind profile based on neural network using SODAR data at Guarulhos Airport,Brazil

A generalized regression neural network model was tested – as a nowcasting tool – to forecast the low wind profiles up to 45 min (i.e. at heights of 10, 100, 200, and 300 m) at the Guarulhos International Airport, São Paulo, Brazil. A data set representing over 4 years was generated from sonic detection and ranging and surface meteorological station, which registered vertical wind profiles with intervals from 10 m to approximately 500 m in height every 15 min, and surface meteorological variables were collected each minute, respectively. These data were simultaneously used to train, validate, and test the proposed model. The u and v forecasts generated at 300, 200, and 100 m were better than at 10 m, which could certainly be attributable to the surface roughness. In addition, the results also revealed that the performance of the model is time-dependent – decreasing over time – and that this may be correlated with the fact that the neural network is a statistical rather than physical model. The forecasts of wind components u and v are slightly biased (or closely matched to observations) at all heights, and forecast intervals with maximum values have median and average errors equal to 0.070 and ?0.017 ms^?1, respectively. The forecast model’s results were evaluated using the values of four categorical statistics: probability of detection; probability for non-events; bias; and false-alarm ratio, with respectable minimum and maximum values for u wind principal components equal to 0.841, 0.833, 0.159, 0.981 at 10 m for 45-min forecasts and 0.989, 0.987, 0.011, 0.999 at 300 m for 15-min forecasts, respectively.     

Mass deployment of sustainable transportation: evaluation of factors that influence electric vehicle adoption

Mass penetration of electric vehicles into the market will have a number of impacts and benefits, including the ability to substantially reduce greenhouse gas emissions from the transportation sector. Therefore, it is expected that in coming years this technology will progressively penetrate the market. This research presents an analysis of factors that influence electric vehicle adoption by modeling the conditions under which an individual, particularly one with an engineering or technical background, is more or less likely to adopt an electric vehicle. This model is developed by considering demographic determinants as well as behavioral and attitudinal measures that affect individual adoption of the technology. The methodology involves applying logistic regression to provide a good fit and predict the response given explanatory variables. Analyzing these outcomes generates empirical findings that better inform electric vehicle technology and policy development. This study takes into account preferences of potential customers and analyzes how individuals with engineering and technology background differ in electric vehicle adoption considerations compared to the general population. Therefore, this research provides both engineers and policy makers with critical information for developing future electric vehicle technology. The model results show that several factors including willingness to pay for new appealing technology, distance driven, perceptions of electric vehicles as good for the environment, perception of EV speed are statistically significant in influencing willingness to purchase an electric vehicle.     

Effects of willow stands on heavy metal concentrations and top soil properties of infrastructure spoil landfills and dredged sediment-derived sites

The effects of willow stand development on top soil properties of uncontaminated infrastructure spoil landfills (ISL) and contaminated dredged sediment landfills (DSL) were assessed. For the ISL, significant increases in Cd, Zn and organic C levels in the top soil (0-10 cm) were detected more than 20 years after disposal. The increases in Cd and Zn concentrations in the top soil were attributed to leaf-associated metal transfer and leaf fall: the relatively high Cd and Zn concentrations in willow leaves resulted in top soil enrichment for these elements. Higher absolute amounts of Cd, Zn and Mn were taken up and recycled during leaf fall on DSL than on ISL, but did not result in significant differences between top soil and deeper soil (10-30 cm) for the DSL. Direct comparison of top soil development between both types of sites is not possible due to differences in stand age and time since disposal. The DSL were characterised by a higher short-range variance for the Cd, Cr, Cu, Pb and Zn concentrations in the top soil than the ISL. During the first years of ripening and dewatering, significant sulphate leaching occurred in the top soil of the DSL.     

An improved adaptive power line interference canceller for electrocardiography

Power line interference may severely corrupt a biomedical recording. Notch filters and adaptive cancellers have been suggested to suppress this interference. We propose an improved adaptive canceller for the reduction of the fundamental power line interference component and harmonics in electrocardiogram (ECG) recordings. The method tracks. the amplitude, phase, and frequency of all the interference components for power line frequency deviations up to about 4 Hz. A comparison is made between the performance of our method, former adaptive cancellers, and a narrow and a wide notch filter in suppressing the fundamental power line interference component. For this purpose a real ECG signal is corrupted by an artificial power line interference signal. The cleaned signal after applying all methods is compared with the original ECG signal. Our improved adaptive canceller shows a signal-to-power-line-interference ratio for the fundamental component up to 30 dB higher than that produced by the other methods. Moreover, our method is also effective for the suppression of the harmonics of the power line interference.     

A novel model-based hearing compensation design using a gradient-free optimization method

We propose a novel model-based hearing compensation strategy and gradient-free optimization procedure for a learning-based hearing aid design. Motivated by physiological data and normal and impaired auditory nerve models, a hearing compensation strategy is cast as a neural coding problem, and a Neurocompensator is designed to compensate for the hearing loss and enhance the speech. With the goal of learning the Neurocompensator parameters, we use a gradient-free optimization procedure, an improved version of the ALOPEX that we have developed, to learn the unknown parameters of the Neurocompensator. We present our methodology, learning procedure, and experimental results in detail; discussion is also given regarding the unsupervised learning and optimization methods.