Article
Computer Science, Artificial Intelligence
Yi Ding, Neethu Robinson, Su Zhang, Qiuhao Zeng, Cuntai Guan
Summary: TSception is a multi-scale convolutional neural network that can classify emotions from EEG. It learns the temporal dynamics and spatial asymmetry of EEG and achieves higher classification accuracies and F1 scores.
IEEE TRANSACTIONS ON AFFECTIVE COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Michael Teichmann, Rene Larisch, Fred H. Hamker
Summary: Computational neuroscience models of vision and neural network models for object recognition serve different research agendas, but should be validated on object recognition tasks for improved practicality and reliability.
Article
Computer Science, Information Systems
Jawaher Alghamdi, Thair Al-Dala'in
Summary: This study proposes a deep neural network model that can automatically extract salient features for predicting crime categories using real-world crime data from the Chicago open data portal. The results show that the model outperforms the baseline model and other algorithms, with an average improvement of 6% in macro F1 score, indicating its high effectiveness in accurately predicting crime categories.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Birgitta Dresp-Langley
Summary: This selective review explores biologically inspired learning as a model for intelligent robot control and sensing technology, and explains the importance of Hebbian synaptic learning for machine learning and intelligence based on specific examples. The potential of adaptive learning and unsupervised control is discussed in relation to the generation of functional complexity.
Article
Chemistry, Multidisciplinary
Hye-jin Shim, Jee-weon Jung, Ju-ho Kim, Ha-jin Yu
Summary: This study investigates misclassified pairs of classes in acoustic scene classification that share similar properties, proposing a max feature map method to replace non-linear activation functions and exploring data augmentation and deep architecture modules to enhance discriminative power and mitigate overfitting. Experiments using a specific dataset demonstrate that the proposed system consistently outperforms the baseline, achieving an accuracy of 70.4% compared to the baseline's 65.1%.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Artificial Intelligence
Bo-Yu Tsai, Sandeep Vara Sankar Diddi, Li-Wei Ko, Shuu-Jiun Wang, Chi-Yuan Chang, Tzyy-Ping Jung
Summary: Brain-computer interface (BCI) is a technology that translates brain signals into actions by establishing direct communication between the brain and external devices. This study proposes an adaptive online technique to remove artifacts from electroencephalography (EEG) recordings, improving BCI performance.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Dawei Cheng, Zhibin Niu, Liqing Zhang
Summary: Small- and medium-sized enterprises (SMEs) can enhance their financial security by guaranteeing each other and obtaining loans from commercial banks. However, this may introduce default risk, especially when many SMEs form complex networks. To address this, macroprudential oversight of the guarantee network is crucial to prevent potential financial crises.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Gabriele Lagani, Fabrizio Falchi, Claudio Gennaro, Giuseppe Amato
Summary: This paper investigates the application of Hebbian learning strategies in training Convolutional Neural Networks (CNNs). Experimental comparisons are made between Hebbian learning and other methods, demonstrating the effectiveness of Hebbian learning in training feature extraction layers and reducing training time.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Physics, Fluids & Plasmas
Hyojun A. Lee, Luiz G. A. Alves, Luis A. Nunes Amaral
Summary: This study compares transmission dynamics between temporal multipartite networks and their time-aggregated unipartite projections, and finds significant differences on three levels. The ratio of the number of nodes to the number of active edges over the time-aggregation scale determines the ability of projected networks to capture the dynamics on multipartite networks.
Article
Chemistry, Analytical
Mircea Hulea, George Iulian Uleru, Constantin Florin Caruntu
Summary: The control of anthropomorphic hands should be carried out by artificial units with high biological plausibility, such as adaptive spiking neural networks. These networks can enable robots to learn motions independently through mechanisms like Hebbian learning. This bioinspired concept allows robots to stop their movements based on specific angles without the need for external stimuli.
Article
Computer Science, Information Systems
William L. Croft, Jorg-Rudiger Sack, Wei Shi
Summary: This paper discusses the privacy protection issues of obfuscating facial images, proposes a new method to address the deficiencies of obfuscation, while ensuring realistic output and preserving useful features in the images. Additionally, it achieves differential privacy for any image and suggests a method to modify pixel intensities without the need for a trained model.
PEER-TO-PEER NETWORKING AND APPLICATIONS
(2021)
Article
Neurosciences
Chengcheng Fan, Banghua Yang, Xiaoou Li, Peng Zan
Summary: This study proposed a method to enhance the decoding accuracy of EEG signals in brain-computer interface. The method generated temporal segments using a sliding window strategy and extracted temporal, frequency, and phase features from different segments. These features were stacked into 3D feature maps, and a compact 3D-CNN model was designed to efficiently extract them. The proposed method achieved good classification accuracy in different tasks and datasets.
FRONTIERS IN NEUROSCIENCE
(2023)
Article
Computer Science, Artificial Intelligence
Abhinav Rai, Fadime Sener, Angela Yao
Summary: Modeling the visual changes brought by an action to a scene is crucial for understanding videos. The current approach of processing one local neighborhood at a time limits the learning of contextual relationships over longer ranges. To address this, we propose TROI, a module that enables CNNs to reason between separated mid-level feature representations in space and time.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2022)
Article
Computer Science, Information Systems
Yingji Li, Yue Wu, Mingchen Sun, Bo Yang, Ying Wang
Summary: This study proposes a Transformer-based continuous dynamic network representation learning model called T-TGNN, which aggregates global information in continuous dynamic networks. By modeling the temporal changes in networks using ordinary neural differential equations and utilizing Transformer mechanisms to aggregate temporal and structural information, this model can better capture the evolution of dynamic networks.
INFORMATION SCIENCES
(2023)
Article
Materials Science, Multidisciplinary
Haiyang Song, Yinghe Zhao, Eleanor Turner, Yu Wu, Yuan Li, Menghao Wu, Guang Feng, Huiqiao Li, Tianyou Zhai
Summary: Researchers have established a new framework using the theory of mutual information to capture 2D van der Waals magnets with high probability for experimental demonstration from materials science literature. This framework has the potential to revolutionize the experimental discovery of 2D vdW magnets.
Article
Engineering, Biomedical
F. Mattioli, C. Porcaro, G. Baldassarre
Summary: This study aims to establish communication paths between the brain and external devices through Brain-Computer Interface (BCI) and has shown great potential for methods based on motor imagery (MI). A new approach based on convolutional neural network was proposed to classify five brain states, achieving high accuracy in the experiment.
JOURNAL OF NEURAL ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Giovanni Granato, Gianluca Baldassarre
Summary: Executive functions depend on integrated cortical-basal ganglia brain systems and form the basis of flexible human behavior. Various computational models have been proposed for studying cognitive flexibility, with the Wisconsin card sorting test (WCST) being an important tool in investigating flexibility.
Article
Environmental Sciences
Marco Mirolli, Luca Simione, Monica Martoni, Marco Fabbri
Summary: The study found that the COVID-19 lockdown decreased levels of mindfulness, increased distress and sleep problems, and the effects of lockdown on sleep were fully mediated by mindfulness and distress, with the acceptance component playing a leading role.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2021)
Article
Computer Science, Artificial Intelligence
Vieri Giuliano Santucci, Davide Montella, Gianluca Baldassarre
Summary: When faced with the problem of autonomously learning to achieve multiple goals, researchers typically focus on problems where each goal can be solved using just one policy. However, in environments presenting different contexts, the same goal might require different skills to solve. We propose a novel robotic architecture, C-GRAIL, that can autonomously detect new relevant contexts and quickly learn policies for new contexts using transfer learning techniques.
IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Kristsana Seepanomwan, Daniele Caligiore, Kevin J. O'Regan, Gianluca Baldassarre
Summary: Developmental psychology experiments show that infants' ability to use tools suddenly emerges around 18 months. Using a developmental-robotics model, two hypotheses were proposed and tested to explain this phenomenon. The results indicate that only the hypothesis about intrinsic motivation can explain the sudden improvement in tool use.
IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS
(2022)
Article
Mathematical & Computational Biology
Pierangelo Afferni, Federico Cascino-Milani, Andrea Mattera, Gianluca Baldassarre
Summary: This article introduces a neuro-inspired computational model to explain the mechanism behind the human brain's lifelong learning capability. The model incorporates two key factors - dopamine encoding and homeostatic plasticity mechanism, and shows positive results in experiments.
FRONTIERS IN COMPUTATIONAL NEUROSCIENCE
(2022)
Article
Computer Science, Artificial Intelligence
Marina Di Vincenzo, Francesco Palini, Maria De Marsico, Anna M. Borghi, Gianluca Baldassarre
Summary: This study compares the usability of a new NUI with a traditional interface in controlling a simulated drone. While the specific NUI used led to lower performance, it was evaluated as more natural and embodied. The results call for further investigation into possible improvements of the NUI.
FRONTIERS IN NEUROROBOTICS
(2022)
Article
Biology
Francesco Mannella, Luca Tummolini
Summary: This study proposes a mechanism to support concept formation by initiating learning through online interaction of perception and intentional action. By mapping multi-modal sensory patterns and motor repertoire in the same low-dimensional representation space, the acquisition of these mappings can be mutually constrained by maximizing convergence between sensory and motor representations. The results demonstrate that this intrinsically motivated learning process can create a cross-modal categorization system with semantic content, supporting perception and intentional action selection, and have the ability to re-enact its own multi-modal experiences, thus kick-starting the formation of concepts grounded in the external environment.
PHILOSOPHICAL TRANSACTIONS OF THE ROYAL SOCIETY B-BIOLOGICAL SCIENCES
(2023)
Article
Neurosciences
Marco Fabbri, Luca Simione, Laura Catalano, Marco Mirolli, Monica Martoni
Summary: Attentional bias toward sleep-related stimuli is related to the severity of insomnia symptoms, as indicated by slower reaction times for sleep-related words in individuals with subclinical and moderate/severe sleep groups. Furthermore, only the moderate/severe sleep group showed a significant disengagement from sleep-related information, suggesting greater difficulties in shifting attention away from sleep-related stimuli in insomnia.
Editorial Material
Audiology & Speech-Language Pathology
Anna M. Borghi, Claudia Mazzuca, Angelo Mattia Gervasi, Francesco Mannella, Luca Tummolini
Summary: In his position paper, Calzavarini (2023) argues that recent studies in multisensory research challenge the claim that word meaning is grounded in modality-specific representations. Here we argue that the coherence of the Grounded Cognition model does not require the modality-specificity assumption. In fact, already some strong versions of the model had relied on a multimodal view of the sensorimotor cortex that seems consistent with much of the evidence discussed in the paper. Next, we address some possible consequences for empirical research using behavioural methods, particularly for norming studies on word meaning. We conclude by identifying some open issues.
LANGUAGE COGNITION AND NEUROSCIENCE
(2023)
Article
Computer Science, Artificial Intelligence
Charles Wilmot, Gianluca Baldassarre, Jochen Triesch
Summary: A key competence for open-ended learning is the formation of increasingly abstract representations that drive complex behavior. In a multimodal setting, generic lossy compression of sensory input naturally extracts abstract representations that prioritize shared information across different modalities. The proposed architecture utilizing autoencoder neural networks demonstrates the validity of the approach and its applicability to embodied agents.
IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS
(2023)
Editorial Material
Computer Science, Artificial Intelligence
Kathryn Kasmarik, Gianluca Baldassarre, Vieri Giuliano Santucci, Jochen Triesch
IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS
(2023)
Meeting Abstract
Psychology, Multidisciplinary
Dara G. Friedman-Wheeler, Alessandro Giannandrea, Monica Martoni, Sara Schairer, Marco Mirolli, Hope Chang, Jessica Lombardi, Alex Tribo
ANNALS OF BEHAVIORAL MEDICINE
(2022)
Article
Clinical Neurology
Marco Fabbri, Luca Simione, Monica Martoni, Marco Mirolli
Summary: Several studies have shown that the COVID-19 pandemic negatively affects sleep quality and mood, but the underlying mechanisms are not well understood. Recent research indicates that the acceptance aspect of mindfulness can reduce anxiety, leading to improved sleep quality. This cross-sectional study aimed to examine changes in mindfulness traits, sleep-wake quality, and general distress before, during, and after the first wave of COVID-19, and to test a model in which acceptance influences sleep through anxiety during each period.
Article
Psychology, Clinical
Luca Simione, Antonino Raffone, Marco Mirolli
Summary: The study found little support for the two tested MAT tenets, and instead strongly supported the alternative view that the beneficial effects of mindfulness on psychological outcomes depend mostly on acceptance.