Perceptual learning, roving and the unsupervised bias

Perceptual learning improves perception through training. Perceptual learning improves with most stimulus types but fails when . certain stimulus types are mixed during training (roving). This result is surprising because classical supervised and unsupervised neural network models can cope easily with roving conditions. What makes humans so inferior compared to these models? As experimental and conceptual work has shown, human perceptual learning is neither supervised nor unsupervised but reward-based learning. Reward-based learning suffers from the so-called unsupervised bias, i.e., to prevent synaptic " drift" , the . average reward has to be exactly estimated. However, this is impossible when two or more stimulus types with different rewards are presented during training (and the reward is estimated by a running average). For this reason, we propose no learning occurs in roving conditions. However, roving hinders perceptual learning only for combinations of similar stimulus types but not for dissimilar ones. In this latter case, we propose that a critic can estimate the reward for each stimulus type separately. One implication of our analysis is that the critic cannot be located in the visual system. © 2011 Elsevier Ltd.

Perceptual thresholds for acoustical guitar models

Synthesised acoustic guitar sounds based on a detailed physical model are used to provide input for psychoacoustical testing. Thresholds of perception are found for changes in the main parameters of the model. Using a three-alternative forced-choice procedure, just-noticeable differences are presented for changes in frequency and damping of the modes of the guitar body, and also for changes in the tension, bending stiffness and damping parameters of the strings. These are compared with measured data on the range of variation of these parameters in a selection of guitars. © S. Hirzel Verlag © EAA.

MCBoost: Multiple classifier boosting for perceptual co-clustering of images and visual features

We present a new co-clustering problem of images and visual features. The problem involves a set of non-object images in addition to a set of object images and features to be co-clustered. Co-clustering is performed in a way that maximises discrimination of object images from non-object images, thus emphasizing discriminative features. This provides a way of obtaining perceptual joint-clusters of object images and features. We tackle the problem by simultaneously boosting multiple strong classifiers which compete for images by their expertise. Each boosting classifier is an aggregation of weak-learners, i.e. simple visual features. The obtained classifiers are useful for object detection tasks which exhibit multimodalities, e.g. multi-category and multi-view object detection tasks. Experiments on a set of pedestrian images and a face data set demonstrate that the method yields intuitive image clusters with associated features and is much superior to conventional boosting classifiers in object detection tasks.

Effect of bandgap narrowing on performance of modern power devices

The effect of the bandgap narrowing (BGN) on performance of power devices is investigated in detail in this paper. The analysis reveals that the change in the energy band structure caused by BGN can strongly affect the conductivity modulation of the bipolar devices resulting in a completely different performance. This is due to the modified injection efficiency under high-level injection conditions. Using a comprehensive analysis of the injection efficiency in a p-n junction, an analytical model for this phenomenon is developed. BGN model tuning has been proved to be essential in accurately predicting the performance of a lateral insulated-gate bipolar transistor (IGBT). Other devices such as p-i-n diodes or punch-through IGBTs are significantly affected by the BGN, while others, such as field-stop IGBTs or power MOSFETs, are only marginally affected. © 2013 IEEE.

Theoretical study of line narrowing in a sub-terahertz monolithic dual laser source with an integrated reflector

A monolithic design is proposed for low-noise sub-THz signal generation by integrating a reflector onto a dual laser source. The reflectivity and the position of such a reflector can be adjusted to obtain constructive feedback from the reflector to both lasers, thus causing a Vernier feedback effect. As a result, 10-fold line narrowing, the narrowing being limited by the resolution of the simulation, is predicted using a transmission line model. Finally, a simple control scheme using an electrical feedback loop to adjust laser biases is proposed to maintain the line narrowing performance. This line narrowing technique, comprising a passive integrated reflector, could allow the development of a low-cost, compact and energy-efficient solution for high-purity sub-THz signal generation. © The Institution of Engineering and Technology 2014.