Sousa, Iam Palatnik deVellasco, Marley M. B. R.Silva, Eduardo Costa da
12页
查看更多>>摘要：A novel evolutionary approach for Explainable Artificial Intelligence is presented: the "Evolved Expla-nations "model (EvEx). This methodology combines Local Interpretable Model Agnostic Explanations (LIME) with Multi-Objective Genetic Algorithms to allow for automated segmentation parameter tuning in image classification tasks. In this case, the dataset studied is Patch-Camelyon, comprised of patches from pathology whole slide images. A publicly available Convolutional Neural Network (CNN) was trained on this dataset to provide a binary classification for presence/absence of lymph node metastatic tissue. In turn, the classifications are explained by means of evolving segmentations, seeking to optimize three evaluation goals simultaneously. The final explanation is computed as the mean of all explanations generated by Pareto front individuals, evolved by the developed genetic algorithm. To enhance reproducibility and traceability of the explanations, each of them was generated from several different seeds, randomly chosen. The observed results show remarkable agreement between different seeds. Despite the stochastic nature of LIME explanations, regions of high explanation weights proved to have good agreement in the heat maps, as computed by pixel-wise relative standard deviations. The found heat maps coincide with expert medical segmentations, which demonstrates that this methodology can find high quality explanations (according to the evaluation metrics), with the novel advantage of automated parameter fine tuning. These results give additional insight into the inner workings of neural network black box decision making for medical data.(c) 2021 Elsevier Ltd. All rights reserved.
Li, JingjingLiu, YijunDong, XianlingSaripan, M. Iqbal...
14页
查看更多>>摘要：This study aims to explore an effective method to evaluate spatial cognitive ability, which can effectively extract and classify the feature of EEG signals collected from subjects participating in the virtual reality (VR) environment; and evaluate the training effect objectively and quantitatively to ensure the objectivity and accuracy of spatial cognition evaluation, according to the classification results. Therefore, a multi-dimensional conditional mutual information (MCMI) method is proposed, which could calculate the coupling strength of two channels considering the influence of other channels. The coupled characteristics of the multi-frequency combination were transformed into multi spectral images, and the image data were classified employing the convolutional neural networks (CNN) model. The experimental results showed that the multi-spectral image transform features based on MCMI are better in classification than other methods, and among the classification results of six band combinations, the best classification accuracy of Beta1-Beta2-Gamma combination is 98.3%. The MCMI characteristics on the Beta1-Beta2-Gamma band combination can be a biological marker for the evaluation of spatial cognition. The proposed feature extraction method based on MCMI provides a new perspective for spatial cognitive ability assessment and analysis. (C)& nbsp;2021 Elsevier Ltd. All rights reserved.
查看更多>>摘要：Fora class of quaternion-valued neural networks (QVNNs) with discrete and distributed time delays, its finite-time synchronization (FTSYN) is addressed in this paper. Instead of decomposition, a direct analytical method named two-step analysis is proposed. That method can always be used to study FTSYN, under either 1-norm or 2-norm of quaternion. Compared with the decomposing method, the two-step method is also suitable for models that are not easily decomposed. Furthermore, a switching controller based on the two-step method is proposed. In addition, two criteria are given to realize the FTSYN of QVNNs. At last, three numerical examples illustrate the feasibility, effectiveness and practicability of our method.(c) 2021 Elsevier Ltd. All rights reserved.
查看更多>>摘要：Recent studies have shown that the choice of activation function can significantly affect the performance of deep learning networks. However, the benefits of novel activation functions have been inconsistent and task dependent, and therefore the rectified linear unit (ReLU) is still the most commonly used. This paper proposes a technique for customizing activation functions automatically, resulting in reliable improvements in performance. Evolutionary search is used to discover the general form of the function, and gradient descent to optimize its parameters for different parts of the network and over the learning process. Experiments with four different neural network architectures on the CIFAR-10 and CIFAR-100 image classification datasets show that this approach is effective. It discovers both general activation functions and specialized functions for different architectures, consistently improving accuracy over ReLU and other activation functions by significant margins. The approach can therefore be used as an automated optimization step in applying deep learning to new tasks. (c) 2022 Elsevier Ltd. All rights reserved.
查看更多>>摘要：This article mainly dedicates on the issue of finite-time stabilization of complex-valued neural networks with proportional delays and inertial terms via directly constructing Lyapunov functions without separating the original complex-valued neural networks into two real-valued subsystems equivalently. First of all, in order to facilitate the analysis of the second-order derivative caused by the inertial term, two intermediate variables are introduced to transfer complex-valued inertial neural networks (CVINNs) into the first-order differential equation form. Then, under the finite-time stability theory, some new criteria with less conservativeness are established to ensure the finite-time stabilizability of CVINNs by a newly designed complex-valued feedback controller. In addition, for reducing expenses of the control, an adaptive control strategy is also proposed to achieve the finite time stabilization of CVINNs. At last, numerical examples are given to demonstrate the validity of the derived results.(C) 2022 Elsevier Ltd. All rights reserved.
查看更多>>摘要：Metric learning has attracted a lot of interest in classification tasks due to its efficient performance. Most traditional metric learning methods are based on k-nearest neighbors (kNN) classifiers to make decisions, while the choice k affects the generalization. In this work, we propose an end-to-end metric learning framework. Specifically, a new linear metric learning (LMML) is first proposed to jointly learn adaptive metrics and the optimal classification hyperplanes, where dissimilar samples are separated by maximizing classification margin. Then a nonlinear metric learning model (called RLMML) is developed based on a bound nonlinear kernel function to extend LMML. The non-convexity of the proposed models makes them difficult to optimize. The half-quadratic optimization algorithms are developed to solve iteratively the problems, by which the optimal classification hyperplane and adaptive metric are alternatively optimized. Moreover, the resulting algorithms are proved to be convergent theoretically. Numerical experiments on different types of data sets show the effectiveness of the proposed algorithms. Finally, the Wilcoxon test shows also the feasibility and effectiveness of the proposed models. (C)& nbsp;2022 Elsevier Ltd. All rights reserved.
查看更多>>摘要：Transformer-based architectures have shown great success in image captioning, where self-attention module can model source and target interaction (e.g., object-to-object, object-to-word, word-to-word). However, the global information is not explicitly considered in the attention weight calculation, which is essential to understand the scene content. In this paper, we propose Dual Global Enhanced Transformer (DGET) to incorporate global information in the encoding and decoding stages. Concretely, in DGET, we regard the grid feature as the visual global information and adaptively fuse it into region features in each layer by a novel Global Enhanced Encoder (GEE). During decoding, we proposed Global Enhanced Decoder (GED) to explicitly utilize the textual global information. First, we devise the context encoder to encode the existing caption generated by classic captioner as a context vector. Then, we use the context vector to guide the decoder to generate accurate words at each time step. To validate our model, we conduct extensive experiments on the MS COCO image captioning dataset and achieve superior performance over many state-of-the-art methods.(c) 2022 Elsevier Ltd. All rights reserved.
查看更多>>摘要：Currently, signed network representation has been applied to many fields, e.g., recommendation platforms. A mainstream paradigm of network representation is to map nodes onto a low-dimensional space, such that the node proximity of interest can be preserved. Thus, a key aspect is the node proximity evaluation. Accordingly, three new node proximity metrics were proposed in this study, based on the rigorous theoretical investigation on a new distance metric signed average first passage time (SAFT). SAFT derives from a basic random-walk quantity for unsigned networks and can capture high-order network structure and edge signs. We conducted network representation using the proposed proximity metrics and empirically exhibited our advantage in solving two downstream tasks - sign prediction and link prediction. The code is publicly available. (C)& nbsp;2022 Elsevier Ltd. All rights reserved.
查看更多>>摘要：To explain the working mechanism of ResNet and its variants, this paper proposes a novel argument of shallow subnetwork first (SSF), essentially low-degree term first (LDTF), which also applies to the whole neural network family. A neural network with shortcut connections behaves as an ensemble of a number of subnetworks of differing depths. Among the subnetworks, the shallow subnetworks are trained firstly, having great effects on the performance of the neural network. The shallow subnetworks roughly correspond to low-degree polynomials, while the deep subnetworks are opposite. Based on Taylor expansion, SSF is consistent with LDTF. ResNet is in line with Taylor expansion: shallow subnetworks are trained firstly to keep low-degree terms, avoiding overfitting; deep subnetworks try to maintain high-degree terms, ensuring high description capacity. Experiments on ResNets and DenseNets show that shallow subnetworks are trained firstly and play important roles in the training of the networks. The experiments also reveal the reason why DenseNets outperform ResNets: The subnetworks playing vital roles in the training of the former are shallower than those in the training of the latter. Furthermore, LDTF can also be used to explain the working mechanism of other ResNet variants (SE-ResNets and SK-ResNets), and the common phenomena occurring in many neural networks. (C)& nbsp;& nbsp;2022 Elsevier Ltd. All rights reserved.
NSTL
Elsevier