Search

In this paper, we believe that the mixed selectivity of neuron in the top layer encodes distributed information produced from other neurons to offer a significant computational advantage over recognition accuracy. Thus, this paper proposes a hierarchical network framework that the learning behaviors of features combined from hundreds of midlayers. First, a subnetwork neuron, which itself could be constructed by other nodes, is functional as a subspace features extractor. The top layer of a hierarchical network needs subspace features produced by the subnetwork neurons to get rid of factors that are not relevant, but at the same time, to recast the subspace features into a mapping space so that the hierarchical network can be processed to generate more reliable cognition. Second, this paper shows that with noniterative learning strategy, the proposed method has a wider and shallower structure, providing a significant role in generalization performance improvements. Hence, compared with other state-of-the-art methods, multiple channel features with the proposed method could provide a comparable or even better performance, which dramatically boosts the learning speed. Our experimental results show that our platform can provide a much better generalization performance than 55 other state-of-the-art methods.

It is clear that the learning effectiveness and learning speed of neural networks are in general far slower than required, which has been a major bottleneck for many applications. Recently, a simple and efficient learning method, referred to as extreme learning machine (ELM), was proposed by Huang , which has shown that, compared to some conventional methods, the training time of neural networks can be reduced by a thousand times. However, one of the open problems in ELM research is whether the number of hidden nodes can be further reduced without affecting learning effectiveness. This brief proposes a new learning algorithm, called bidirectional extreme learning machine (B-ELM), in which some hidden nodes are not randomly selected. In theory, this algorithm tends to reduce network output error to 0 at an extremely early learning stage. Furthermore, we find a relationship between the network output error and the network output weights in the proposed B-ELM. Simulation results demonstrate that the proposed method can be tens to hundreds of times faster than other incremental ELM algorithms.

Artificial neural network training algorithms aim to optimize the network parameters regarding the pre-defined cost function. Gradient-based artificial neural network training algorithms support iterative learning and have gained immense popularity for training different artificial neural networks end-to-end. However, training through gradient methods is time-consuming. Another family of training algorithms is based on the Moore–Penrose inverse, which is much faster than many other gradient methods. Nevertheless, most of those algorithms are non-iterative and thus do not support mini-batch learning in nature. This work extends two non-iterative Moore–Penrose inverse-based training algorithms to enable online sequential learning: a single-hidden-layer autoencoder training algorithm and a sub-network-based classifier training algorithm. We further present an approach that uses the proposed autoencoder for self-supervised dimension reduction and then uses the proposed classifier for supervised classification. The experimental results show that the proposed approach achieves satisfactory classification accuracy on many benchmark datasets with extremely low time consumption (up to 50 times faster than the support vector machine on CIFAR 10 dataset). [ABSTRACT FROM AUTHOR]

Blood viscosity is one of the important parameters to characterize hemorheological properties of the human body. Its real-time and dynamic measurement has important physiological significance for studying the development and prevention of chronic diseases. This study researched noninvasive and personalized measurement of microvascular blood viscosity. In the microcirculation capillary network blood flow model, combined with pulse wave parameters, multiple regression analysis was used to fit the simulated radius of personalized physiological blood vessels to calculate the microvascular blood viscosity. The fitted value related to the simulated radius of the physiological blood vessel had a high correlation with the corresponding theoretically derived value (correlation coefficient: 0.904, P ≤ 0.001). The calculated value of the microvascular blood viscosity had a certain correlation with the clinical whole blood viscosity at a low shear rate (correlation coefficient: 0.443, P < 0.05). This algorithm could provide effective means for noninvasive and long-term individual monitoring and family health care. [ABSTRACT FROM AUTHOR]

Rail inspection system (RIS) remains an emergent instrumentation for railway transportation, with its capacity of measuring surface defect on steel rail. However, detecting technique and interpretation of RIS constitute a challenging problem since traditional technologies are expensive and prone to errors. In this paper, a deep multimodel RIS (DM-RIS) is established for surface defect where fast and robust spatially constrained Gaussian mixture model is presented for segmentation proposal and Faster RCNN is utilized for objective location in a parallel structure. First, we incorporate spatial information between pixels into an improved Gaussian mixture model based on Markov random field (MRF) for accurate and rapid defect edge segmentation. Specifically, a direct parameter-learning in expectation–maximization (EM) algorithm is proposed. Meanwhile, to remove nondefect, numerous labeled samples with weak illumination, inequality reflection, external noise, rust, and greasy dirt are fed into Faster RCNN so that DM-RIS is robust environmentally to various light, angle, background, and acquisition equipment. Finally, the joint hit area refers to a real defect. The experimental results demonstrate that the proposed method performs well with 96.74% precision, 94.13% recall, 95.18% overlap, and 0.485 s/frame speed on average, and is robust compared with the related well-established approaches. [ABSTRACT FROM AUTHOR]

Abstract: Much research on Artificial Intelligence (AI) has been focusing on exploring various potential applications of intelligent systems. In most cases, the researches attempt to model human intelligence by mimicking the brain structure and function, but they ignore an important aspect in human learning and decision making: the artificial emotion. In this paper, we present a new unconstrained global optimization method, hybrid chaos optimization algorithm with artificial emotion (HCOAAE), which avoids trapping to local minima, and improves convergence in large space and high-dimension optimization problems. The main purpose of artificial emotion is to mimic decision making behavior process of humans, to choose most suitable parameters of HCOAAE and decide whether to change current search strategy or not in the next iteration. Numerical simulations of 13 benchmark functions with different dimensions are used to test the performance of HCOAAE. Experimental results show that the proposed method significantly outperforms the existing methods in terms of convergence speed, computational effectiveness, and numerical stability. [Copyright &y& Elsevier]