Cognitive Computation最新文献

Abstract

Driver drowsiness is a significant concern and one of the leading causes of traffic accidents. Advances in cognitive neuroscience and computer science have enabled the detection of drivers’ drowsiness using Brain-Computer Interfaces (BCIs) and Machine Learning (ML). However, the literature lacks a comprehensive evaluation of drowsiness detection performance using a heterogeneous set of ML algorithms, being also necessary to study the performance of scalable ML models suitable for groups of subjects. To address these limitations, this work presents an intelligent framework employing BCIs and features based on electroencephalography for detecting drowsiness in driving scenarios. The SEED-VIG dataset is used to evaluate the best-performing models for individual subjects and groups. Results show that Random Forest (RF) outperformed other models used in the literature, such as Support Vector Machine (SVM), with a 78% f1-score for individual models. Regarding scalable models, RF reached a 79% f1-score, demonstrating the effectiveness of these approaches. This publication highlights the relevance of exploring a diverse set of ML algorithms and scalable approaches suitable for groups of subjects to improve drowsiness detection systems and ultimately reduce the number of accidents caused by driver fatigue. The lessons learned from this study show that not only SVM but also other models not sufficiently explored in the literature are relevant for drowsiness detection. Additionally, scalable approaches are effective in detecting drowsiness, even when new subjects are evaluated. Thus, the proposed framework presents a novel approach for detecting drowsiness in driving scenarios using BCIs and ML.

Image hazing poses a significant challenge in various computer vision applications, degrading the visual quality and reducing the perceptual clarity of captured scenes. The proposed AGD-Net utilizes a U-Net style architecture with an Attention-Guided Dense Inception encoder-decoder framework. Unlike existing methods that heavily rely on synthetic datasets which are based on CARLA simulation, our model is trained and evaluated exclusively on realistic data, enabling its effectiveness and reliability in practical scenarios. The key innovation of AGD-Net lies in its attention-guided mechanism, which empowers the network to focus on crucial information within hazy images and effectively suppress artifacts during the dehazing process. The dense inception modules further advance the representation capabilities of the model, facilitating the extraction of intricate features from the input images. To assess the performance of AGD-Net, a detailed experimental analysis is conducted on four benchmark haze datasets. The results show that AGD-Net significantly outperforms the state-of-the-art methods in terms of PSNR and SSIM. Moreover, a visual comparison of the dehazing results further validates the superior performance gains achieved by AGD-Net over other methods. By leveraging realistic data exclusively, AGD-Net overcomes the limitations associated with synthetic datasets which are based on CARLA simulation, ensuring its adaptability and effectiveness in real-world circumstances. The proposed AGD-Net offers a robust and reliable solution for single-image dehazing, presenting a significant advancement over existing methods.

Working memory (WM) is a crucial cognitive function required to maintain and manipulate information that is no longer present through the senses. Two key features of WM are its limited capacity and the emergence of serial order effects. This study investigates how synaptic facilitation and diverse display dynamics influence the encoding and retention of multiple items in WM. A biophysically inspired attractor model of WM, endowed with synaptic facilitation, is considered in this study. The investigation delves into the behaviour of the model under both sequential and simultaneous display protocols. Synaptic facilitation plays a crucial role in establishing the response of the WM system by regulating resource allocation during the encoding stage. It boosts WM capacity and is a key mechanism in the emergence of serial order effects. The synaptic facilitation time constant ((tau _F)) is critical in modulating these effects, and its heterogeneity in the prefrontal cortex (PFC) may contribute to the combination of primacy and recency effects observed experimentally. Additionally, we demonstrate that the WM capacity exhibited by the network is heavily influenced by factors such as the stimuli nature, and their display duration. Although the network connectivity determines the WM capacity by regulating the excitation-inhibition balance, the display protocol modulates its effective limit. Our findings shed light on how different stimulation protocol dynamics affect WM, underscoring the importance of synaptic facilitation and experimental protocol design in modulating WM capacity.

In recent times, there has been active research on multi-disease classification that aim to diagnose lung diseases and respiratory conditions using respiratory data. Recorded respiratory data can be used to diagnose various chronic diseases, such as asthma and pneumonia by applying different feature extraction methods. Previous studies have primarily focused on respiratory disease classification using 2D image conversion techniques, such as spectrograms and mel frequency cepstral coefficients (MFCC) for respiratory data. However, as the number of respiratory disease classes increased, the classification accuracy tended to decrease. To address this challenge, this study proposes a novel approach that combines 1D and 2D data to enhance the multi-classification performance regarding respiratory disease. We incorporated widely used 2D representations such as spectrograms, gammatone-based spectrograms, and MFCC images, along with raw data. The proposed respiratory disease classification method comprises 2D data conversion, combined data generation, classification model development, and multi-disease classification steps. Our method achieved high classification accuracies of 92.93%, 91.30%, and 88.58% using the TCN, Wavenet, and BiLSTM models, respectively. Compared to using solely 1D data, our approach demonstrated a 4.89% improvement in accuracy and more than 3 times better training speed when using only 2D data. These results confirmed the superiority of the proposed method. This allows us to leverage the advantages of fast learning provided by time-series models, as well as the high classification accuracy demonstrated by 2D image approaches.

Because the abstracts contain complex information and the labels of abstracts do not contain information about categories, it is difficult for cognitive models to extract comprehensive features to match the corresponding labels. In this paper, a cognitively inspired multi-granularity model incorporating label information (LIMG) is proposed to solve these problems. Firstly, we use information of abstracts to give labels the actual semantics. It can improve the semantic representation of word embeddings. Secondly, the model uses the dual channel pooling convolutional neural network (DCP-CNN) and the timescale shrink gated recurrent units (TSGRU) to extract multi-granularity information of abstracts. One of the channels in DCP-CNN highlights the key content and the other is used for TSGRU to extract context-related features of abstracts. Finally, TSGRU adds a timescale to retain the long-term dependence by recuring the past information and a soft thresholding algorithm to realize the noise reduction. Experiments were carried out on four benchmark datasets: Arxiv Academic Paper Dataset (AAPD), Web of Science (WOS), Amazon Review and Yahoo! Answers. As compared to the baseline models, the accuracy is improved by up to 3.36%. On AAPD (54,840 abstracts) and WOS (46,985 abstracts) datasets, the micro-F1 score reached 75.62% and 81.68%, respectively. The results show that acquiring label semantics from abstracts can enhance text representations and multi-granularity feature extraction can inspire the cognitive system’s understanding of the complex information in abstracts.

The psychological factors of experts play a special role in the process of decision-making, especially in some situations that experts are not completely rational. Traditional decision-making methods always just focus on the aggregation of positive preference information, which do not take the negative attribute information into account at the same time. The probabilistic dual hesitant fuzzy set (PDHFS) is one of the latest fuzzy sets, which can depict experts’ positive and negative preference information with the corresponding probability at the same time. Therefore, to manage the applications with incomplete rationality and two opposite kinds of uncertain preference information, this paper considers the influence of psychological behavior on decision-making results and introduces an interactive method based on the prospect theory. Taking the advantages of PDHFSs in group decision-making problems, we propose the distance measure of PDHFSs, based on which an improved TODIM (TOmada deDecisão Iterativa Multicritério) method under the probabilistic dual hesitant fuzzy environment is also developed. Meanwhile, we provide the specific implementation process of the proposed method. The proposed improved TODIM is applied to the risk evaluation of Arctic geopolitics. We also make a comparison with the traditional aggregation method of PDHFSs. The difference among alternatives obtained by the proposed TODIM method with prospect theory is much greater than the traditional aggregation methods without prospect theory. This paper highlights the benefits and advantages of the proposed TODIM method that is developed based on the prospect theory and probabilistic dual hesitant fuzzy distance measure.

This paper focuses on the adaptive control issue for a class of uncertain nonlinear systems subject to full state constraints and external disturbance. A novel adaptive nonlinear observer is proposed to compensate for disturbance variables in the transformed system. Combining with radial basis function neural networks (RBFNNs) and nonlinear mapping (NM) mechanism, the constrained system is transformed into an unconstrained form and the system uncertainties are effectively handled. Besides that, an adaptive tracking control approach is formulated by invoking backstepping techniques and the event-sampled scheme is utilized to address the sparsity of resources. The adaptive control problem can be addressed with the proposed algorithm, applying the Lyapunov functions, RBF NNs theory, and inequality techniques. Based on the Lyapunov stability theory, it is proved that the system can never violate the specified state constraints and all the closed-loop signals are semiglobally uniformly ultimately bounded (SGUUB). The validity of the proposed approach is well illustrated by a developed numerical example.

In realistic sparse reward tasks, existing theoretical methods cannot be effectively applied due to the low sampling probability ofrewarded episodes. Profound research on methods based on intrinsic rewards has been conducted to address this issue, but exploration with sparse rewards remains a great challenge. This paper describes the loop enhancement effect in exploration processes with sparse rewards. After each fully trained iteration, the execution probability of ineffective actions is higher than thatof other suboptimal actions, which violates biological habitual behavior principles and is not conducive to effective training. This paper proposes corresponding theorems of relieving the loop enhancement effect in the exploration process with sparse rewards and a heuristic exploration method based on action effectiveness constraints (AEC), which improves policy training efficiency by relieving the loop enhancement effect. Inspired by the fact that animals form habitual behaviors and goal-directed behaviors through the dorsolateral striatum and dorsomedial striatum. The function of the dorsolateral striatum is simulated by an action effectiveness evaluation mechanism (A2EM), which aims to reduce the rate of ineffective samples and improve episode reward expectations. The function of the dorsomedial striatum is simulated by an agent policy network, which aims to achieve task goals. The iterative training of A2EM and the policy forms the AEC model structure. A2EM provides effective samples for the agent policy; the agent policy provides training constraints for A2EM. The experimental results show that A2EM can relieve the loop enhancement effect and has good interpretability and generalizability. AEC enables agents to effectively reduce the loop rate in samples, can collect more effective samples, and improve the efficiency of policy training. The performance of AEC demonstrates the effectiveness of a biological heuristic approach that simulates the function of the dorsal striatum. This approach can be used to improve the robustness of agent exploration with sparse rewards.

While graph convolution-based methods have become the de-facto standard for graph representation learning, their applications to disease prediction tasks remain quite limited, particularly in the classification of neurodevelopmental and neurodegenerative brain disorders. In this paper, we introduce an aggregator normalization graph convolutional network by leveraging aggregation in graph sampling, as well as skip connections and identity mapping. The proposed model learns discriminative graph node representations by incorporating both imaging and non-imaging features into the graph nodes and edges, respectively, with the aim of augmenting predictive capabilities and providing a holistic perspective on the underlying mechanisms of brain disorders. Skip connections enable the direct flow of information from the input features to later layers of the network, while identity mapping helps maintain the structural information of the graph during feature learning. We benchmark our model against several recent baseline methods on two large datasets, Autism Brain Imaging Data Exchange (ABIDE) and Alzheimer’s Disease Neuroimaging Initiative (ADNI), for the prediction of autism spectrum disorder and Alzheimer’s disease, respectively. Experimental results demonstrate the competitive performance of our approach in comparison with recent baselines in terms of several evaluation metrics, achieving relative improvements of 50% and 13.56% in classification accuracy over graph convolutional networks (GCNs) on ABIDE and ADNI, respectively. Our study involved the development of a graph convolutional aggregation model, which aimed to predict the status of subjects in a population graph. We learned discriminative node representations by utilizing imaging and non-imaging features associated with the graph nodes and edges. Our model outperformed existing graph convolutional-based methods for disease prediction on two large benchmark datasets, as shown through extensive experiments. We achieved significant relative improvements in classification accuracy over GCN and other strong baselines.