ABSTRACT
We show, based on the following three grounds, that the primary visual cortex (V1) is a biological direct-shortcut deep residual learning neural network (ResNet) for sparse visual processing: (1) We first highlight that Gabor-like sets of basis functions, which are similar to the receptive fields of simple cells in the primary visual cortex (V1), are excellent candidates for sparse representation of natural images; i.e., images from the natural world, affirming the brain to be optimized for this. (2) We then prove that the intra-layer synaptic weight matrices of this region can be reasonably first-order approximated by identity mappings, and are thus sparse themselves. (3) Finally, we point out that intra-layer weight matrices having identity mapping as their initial approximation, irrespective of this approximation being also a reasonable first-order one or not, resemble the building blocks of direct-shortcut digital ResNets, which completes the grounds. This biological ResNet interconnects the sparsity of the final representation of the image to that of its intra-layer weights. Further exploration of this ResNet, and understanding the joint effects of its architecture and learning rules, e.g. on its inductive bias, could lead to major advancements in the area of bio-inspired digital ResNets. One immediate line of research in this context, for instance, is to study the impact of forcing the direct-shortcuts to be good first-order approximations of each building block. For this, along with the â 1 -minimization posed on the basis function coefficients the ResNet finally provides at its output, another parallel one could e.g. also be posed on the weights of its residual layers.
Subject(s)
Deep Learning , Visual Perception , Humans , Visual Perception/physiology , Neural Networks, Computer , Primary Visual Cortex/physiology , Models, Neurological , Visual Cortex/physiologyABSTRACT
Background: Ultrasound evaluation of normal, ectopic, asymmetric, and hyperplastic thymus and also its differentiation from abnormalities are challenging in children, and few studies have addressed this issue. This study aimed to investigate the thymus sonographic changes with age. Methods: In this cross-sectional study, 118 healthy children were categorised into six age groups. Sonographic features of the thymus, including volume, anatomical position, symmetry, and echo-texture, were recorded. Results: The thymus was visible at all ages from the suprasternal view. In 77.5% of participants, the thymus gland volume in lobes was symmetrical; however, left (21.2%) and right (1.3%) predominance were also found. The most common position of the thymus was in front of the great vessels (100%) with suprasternal extension (97.5%). The mean volume of thymus was 21.3 ± 10.5 (mm). There was no significant difference in the volumes of the thymus between different age groups. The predominant echo-texture of the thymus in different age groups was hypoechoic with thin echogenic septa (liver-like) in below 2-3 years of age, the appearance of echogenic foci and hyperechoic echo-texture (liver-like with starry sky) in 2-14 years, and uniform hyperechoic echo-texture (fatty liver-like) or geographic echo-texture with coarse reticular pattern in above 14 years. Conclusion: In children, the thymus gland is visible in ultrasound examination in all age groups from the suprasternal view; however, the echo-texture of the normal thymus changes with age. There was no significant correlation between age and sex with total thymic volume. The specificity of these appearances has made ultrasound a problem-solving modality in children.
ABSTRACT
In recent years Dynamic Time Warping (DTW) has emerged as the distance measure of choice for virtually all time series data mining applications. For example, virtually all applications that process data from wearable devices use DTW as a core sub-routine. This is the result of significant progress in improving DTW's efficiency, together with multiple empirical studies showing that DTW-based classifiers at least equal (and generally surpass) the accuracy of all their rivals across dozens of datasets. Thus far, most of the research has considered only the one-dimensional case, with practitioners generalizing to the multi-dimensional case in one of two ways, dependent or independent warping. In general, it appears the community believes either that the two ways are equivalent, or that the choice is irrelevant. In this work, we show that this is not the case. The two most commonly used multi-dimensional DTW methods can produce different classifications, and neither one dominates over the other. This seems to suggest that one should learn the best method for a particular application. However, we will show that this is not necessary; a simple, principled rule can be used on a case-by-case basis to predict which of the two methods we should trust at the time of classification. Our method allows us to ensure that classification results are at least as accurate as the better of the two rival methods, and, in many cases, our method is significantly more accurate. We demonstrate our ideas with the most extensive set of multi-dimensional time series classification experiments ever attempted.