查看更多>>摘要：Recent theoretical and experimental works have connected Hebbian plasticity with the reinforcement learning (RL) paradigm, producing a class of trial-and-error learning in artificial neural networks known as neo-Hebbian plasticity. Inspired by the role of the neuromodulator dopamine in synaptic modification, neo-Hebbian RL methods extend unsupervised Hebbian learning rules with value-based modulation to selectively reinforce associations. This reinforcement allows for learning exploitative behaviors and produces RL models with strong biological plausibility. The review begins with coverage of fundamental concepts in rate-and spike-coded models. We introduce Hebbian correlation detection as a basis for modification of synaptic weighting and progress to neo-Hebbian RL models guided solely by extrinsic rewards. We then analyze state-of-the-art neo-Hebbian approaches to the exploration- exploitation balance under the RL paradigm, emphasizing works that employ additional mechanics to modulate that dynamic. Our review of neo-Hebbian RL methods in this context indicates substantial potential for novel improvements in exploratory learning, primarily through stronger incorporation of intrinsic motivators. We provide a number of research suggestions for this pursuit by drawing from modern theories and results in neuroscience and psychology. The exploration-exploitation balance is a central issue in RL research, and this review is the first to focus on it under the neo-Hebbian RL framework. (C) 2022 Elsevier Ltd. All rights reserved
查看更多>>摘要：Complex-valued neural networks have attracted increasing attention in recent years, while it remains open on the advantages of complex-valued neural networks in comparison with real-valued networks. This work takes one step on this direction by introducing the complex-reaction network with fully-connected feed-forward architecture. We prove the universal approximation property for complex-reaction networks, and show that a class of radial functions can be approximated by a complex-reaction network using the polynomial number of parameters, whereas real-valued networks need at least exponential parameters to reach the same approximation level. For empirical risk minimization, we study the landscape and convergence of complex gradient descents. Our theoretical result shows that the critical point set of complex-reaction networks is a proper subset of that of real-valued networks, which may show some insights on finding the optimal solutions more easily for complex-reaction networks. (C)& nbsp;2022 Elsevier Ltd. All rights reserved.
查看更多>>摘要：Performing pattern analysis over the symmetric positive definite (SPD) manifold requires specific mathematical computations, characterizing the non-Euclidian property of the involved data points and learning tasks, such as the image set classification problem. Accompanied with the advanced neural networking techniques, several architectures for processing the SPD matrices have recently been studied to obtain fine-grained structured representations. However, existing approaches are challenged by the diversely changing appearance of the data points, begging the question of how to learn invariant representations for improved performance with supportive theories. Therefore, this paper designs two Riemannian operation modules for SPD manifold neural network. Specifically, a Riemannian batch regularization (RBR) layer is firstly proposed for the purpose of training a discriminative manifold-to-manifold transforming network with a novelly-designed metric learning regularization term. The second module realizes the Riemannian pooling operation with geometric computations on the Riemannian manifolds, notably the Riemannian barycenter, metric learning, and Riemannian optimization. Extensive experiments on five benchmarking datasets show the efficacy of the proposed approach.(C)& nbsp; 2022 Published by Elsevier Ltd.
查看更多>>摘要：Electroencephalographic measurement of cortical activity subserving motor behavior varies among different individuals, restricting the potential of brain computer interfaces (BCIs) based on motor imagery (MI). How to deal with this variability and thereby improve the accuracy of BCI classification remains a key issue. This paper proposes a deep learning-based approach to transfer the data distribution from BCI-friendly - "golden subjects"to the data from more typical BCI-illiterate users. In this work, we use the perceptual loss to align the dimensionality-reduced BCI-illiterate data with the data of golden subjects in low dimensions, by which a subject transfer neural network (STNN) is proposed. The network consists of two parts: 1) a generator, which generates the transferred BCIilliterate features, and 2) a CNN classifier, which is used for the classification of the transferred features, thus outperforming traditional classification methods both in terms of accuracy and robustness. Electroencephalography (EEG) signals from 25 healthy subjects performing MI of the right hand and foot were classified with an average accuracy of 88.2% +/- 5.1%. The proposed model was further validated on the BCI Competition IV dataset 2b, and was demonstrated to be robust to inter-subject variations. The advantages of STNN allow it to bridge the gap between the golden subjects and the BCI-illiterate ones, paving the way to real-time BCI applications. (c) 2022 Elsevier Ltd. All rights reserved.
查看更多>>摘要：Despite considerable progress in the field of automatic multi-target tracking, several problems such as data association remained challenging. On the other hand, cognitive studies have reported that humans can robustly track several objects simultaneously. Such circumstances happen regularly in daily life, and humans have evolved to handle the associated problems. Accordingly, using braininspired processing principles may contribute to significantly increase the performance of automatic systems able to follow the trajectories of multiple objects. In this paper, we propose a multipleobject tracking algorithm based on dynamic neural field theory which has been proven to provide neuro-plausible processing mechanisms for cognitive functions of the brain. We define several input neural fields responsible for representing previous location and orientation information as well as instantaneous linear and angular speed of the objects in successive video frames. Image processing techniques are applied to extract the critical object features including target location and orientation. Two prediction fields anticipate the objects' locations and orientations in the upcoming frame after receiving excitatory and inhibitory inputs from the input fields in a feed-forward architecture. This information is used in the data association and labeling process. We tested the proposed algorithm on a zebrafish larvae segmentation and tracking dataset and an ant-tracking dataset containing non-rigid objects with spiky movements and frequently occurring occlusions. The results showed a significant improvement in tracking metrics compared to state-of-the-art algorithms.(c) 2022 Elsevier Ltd. All rights reserved.
查看更多>>摘要：The long-tailed distribution in the dataset is one of the major challenges of deep learning. Convolutional Neural Networks have poor performance in identifying classes with only a few samples. For this problem, it has been proved that separating the feature learning stage and the classifier learning stage improves the performance of models effectively, which is called decoupled learning. We use soft labels to improve the performance of the decoupled learning framework by proposing a Dynamic Auxiliary Soft Labels (DaSL) method. Specifically, we design a dedicated auxiliary network to generate auxiliary soft labels for the two different training stages. In the feature learning stage, it helps to learn features with smaller variance within the class, and in the classifier learning stage it helps to alleviate the overconfidence of the model prediction. We also introduce a feature-level distillation method for the feature learning, and improve the learning of general features through multi-scale feature fusion. We conduct extensive experiments on three long-tailed recognition benchmark datasets to demonstrate the effectiveness of our DaSL.(C) 2022 Elsevier Ltd. All rights reserved.
查看更多>>摘要：In the era of big data, there are an increasing number of multisource information data, and multi-view clustering (MVC) algorithms have developed rapidly. However, the affinity matrix learned by most MVC methods is not clean and precise enough and cannot describe the latent structure of multi-view data accurately, which results in poor clustering performance. In this paper, we propose a novel Double Structure Scaled Simplex Representation (DSSSR) method for MVC. Initially, we concatenate the multi-view data into a joint representation. Then, we use the scaled simplex representation (SSR) method on the concatenated data to obtain the affinity matrix. However, the affinity matrix is not clean and precise. Therefore, we use the SSR method again on the obtained affinity matrix to obtain a more accurate and clean affinity matrix. Furthermore, the two-step SSR is integrated into a unified optimization framework, a clean and accurate affinity matrix can be obtained, and the sum of each column vector of the affinity matrix is constrained to be nonnegative and equal to s (0 < s < 1), which can be adjusted to obtain the best clustering performance. Finally, an efficient optimization algorithm based on the augmented Lagrangian method (ALM) for solving the objective function is also designed. The experimental results on some datasets show that this algorithm has better clustering performance than some state-of-the-art algorithms. (C)& nbsp;2022 Elsevier Ltd. All rights reserved.
查看更多>>摘要：The COVID-19 pandemic is an ongoing pandemic and is placing additional burden on healthcare systems around the world. Timely and effectively detecting the virus can help to reduce the spread of the disease. Although, RT-PCR is still a gold standard for COVID-19 testing, deep learning models to identify the virus from medical images can also be helpful in certain circumstances. In particular, in situations when patients undergo routine X-rays and/or CT-scans tests but within a few days of such tests they develop respiratory complications. Deep learning models can also be used for pre-screening prior to RT-PCR testing. However, the transparency/interpretability of the reasoning process of predictions made by such deep learning models is essential. In this paper, we propose an interpretable deep learning model that uses positive reasoning process to make predictions. We trained and tested our model over the dataset of chest CT-scan images of COVID-19 patients, normal people and pneumonia patients. Our model gives the accuracy, precision, recall and F-score equal to 99.48%, 0.99, 0.99 and 0.99, respectively. (C)& nbsp;2022 Elsevier Ltd. All rights reserved.
查看更多>>摘要：A fundamental characteristic of human visual perception is the ability to group together disparate elements in a scene and treat them as a single unit. The mechanisms by which humans create such groupings remain unknown, but grouping seems to play an important role in a wide variety of visual phenomena, and a good understanding of these mechanisms might provide guidance for how to improve machine vision algorithms. Here, we build on a proposal that some groupings are the result of connections in cortical area V2 that join disparate elements, thereby allowing them to be selected and segmented together. In previous instantiations of this proposal, connection formation was based on the anatomy (e.g., extent) of receptive fields, which made connection formation obligatory when the stimulus conditions stimulate the corresponding receptive fields. We now propose dynamic circuits that provide greater flexibility in the formation of connections and that allow for top-down control of perceptual grouping. With computer simulations we explain how the circuits work and show how they can account for a wide variety of Gestalt principles of perceptual grouping, texture segmentation tasks, amodal illusory contours, and ratings of perceived groupings. We propose that human observers use such top-down control to implement task-dependent connection strategies that encourage particular groupings of stimulus elements in order to promote performance on various visual tasks.(C) 2022 Elsevier Ltd. All rights reserved.
查看更多>>摘要：Sparse optimization involving the L-0-norm function as the regularization in objective function has a wide application in many fields. In this paper, we propose a projected neural network modeled by a differential equation to solve a class of these optimization problems, in which the objective function is the sum of a nonsmooth convex loss function and the regularization defined by the L-0-norm function. This optimization problem is not only nonconvex, but also discontinuous. To simplify the structure of the proposed network and let it own better convergence properties, we use the smoothing method, where the new constructed smoothing function for the regularization term plays a key role. We prove that the solution to the proposed network is globally existent and unique, and any accumulation point of it is a critical point of the continuous relaxation model. Except for a special case, which can be easily justified, any critical point is a local minimizer of the considered sparse optimization problem. It is an interesting thing that all critical points own a promising lower bound property, which is satisfied by all global minimizers of the considered problem, but is not by all local minimizers. Finally, we use some numerical experiments to illustrate the efficiency and good performance of the proposed method for solving this class of sparse optimization problems, which include the most widely used models in feature selection of classification learning. (C)& nbsp;2022 Elsevier Ltd. All rights reserved.
NSTL
Elsevier