Article
Psychology, Experimental
Nicholas M. Blauch, Marlene Behrmann, David C. Plaut
Summary: Human unfamiliar face identity perception reflects substantial perceptual expertise, with the advantage for familiar over unfamiliar face identity matching reflecting a learned mapping. The response clarifies the authors' stance on theoretical issues and discusses future research topics in human face recognition and the learning of perceptual representations.
Article
Multidisciplinary Sciences
Tao Yan, Rui Yang, Ziyang Zheng, Xing Lin, Hongkai Xiong, Qionghai Dai
Summary: Photonic neural networks use photons instead of electrons to perform brain-like computations, leading to significantly improved computing performance. However, current architectures are limited to handling data with regular structures and cannot generalize to graph-structured data beyond Euclidean space. In this study, a diffractive graph neural network (DGNN) is proposed to address this limitation by utilizing diffractive photonic computing units (DPUs) and on-chip optical devices. DGNN achieves complex feature representation by capturing dependencies among node neighborhoods during light-speed optical message passing over graph structures. It demonstrates superior performance in node and graph-level classification tasks with benchmark databases, providing a new direction for high-efficiency processing of large-scale graph data structures using deep learning.
Review
Biochemical Research Methods
Shuting Jin, Xiangxiang Zeng, Feng Xia, Wei Huang, Xiangrong Liu
Summary: The increase in biological data and the formation of biomolecule interaction databases have led to the emergence of diverse biological networks, providing valuable resources for understanding biological systems, complex disease discovery, and drug research. However, the complexity of biological networks analysis has also increased with the rise in data volume, necessitating algorithms like deep learning to handle large, heterogeneous, and complex data. Deep learning, with its ability to extract abstract features and process complex graph data structures, is increasingly being used in mining valuable information from network data.
BRIEFINGS IN BIOINFORMATICS
(2021)
Article
Multidisciplinary Sciences
Siddharth Joshi, Brian O. Gallachoir, James Glynn
Summary: This study introduces a novel deep learning neural network architecture called TrebuNet, which mimics the physical process of firing a trebuchet, to estimate transport energy service demand. TrebuNet shows superior performance compared to traditional multivariate linear regression and other state-of-the-art methods when evaluated for regional demand projection in all modes of transport. Additionally, TrebuNet introduces a framework for projecting energy service demand in regions with multiple countries and different socio-economic development pathways, applicable to regression-based tasks with non-uniform variance time series.
SCIENTIFIC REPORTS
(2023)
Article
Computer Science, Information Systems
Seyed Mohammad Ghaffarian, Hamid Reza Shahriari
Summary: This paper explores the utilization of Graph Neural Networks for software vulnerability analysis and proposes an original neural vulnerability analysis approach. Experimental results demonstrate the effectiveness of the proposed method in software vulnerability analysis and answer complementary research questions, including cross-project vulnerability analysis.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Nadezhda Semenova, Laurent Larger, Daniel Brunner
Summary: Deep neural networks have unlocked new applications previously reserved for higher human intelligence, leveraging computing power from special purpose hardware. However, the emulation of neural networks by binary computing leads to unsustainable energy consumption and slow speed. Research shows that noise accumulation in deep neural networks with noisy nonlinear neurons is generally limited, and noise can be completely suppressed when neuron activation functions have a slope smaller than unity.
Article
Physics, Multidisciplinary
Lea Fellner, Marian Kraus, Arne Walter, Frank Duschek, Thomas Bocklitz, Valentina Gabbarini, Riccardo Rossi, Alessandro Puleio, Andrea Malizia, Pasquale Gaudio
Summary: Research on using laser-induced fluorescence (LIF) technology to distinguish organic materials, proposing a simplified experimental design to address interference from other fluorophores in the environment, testing neural networks in identifying different fluorophore signals in mixed samples.
EUROPEAN PHYSICAL JOURNAL PLUS
(2021)
Article
Thermodynamics
Heesoo Shin, Mario Ruettgers, Sangseung Lee
Summary: This study investigates the influence of incorporating spatiotemporal wind data on the performance of wind forecasting neural networks. Convolutional neural networks (CNNs) trained with various scales of spatiotemporal wind data show improved predictive performance. Correlation analyses reveal the impact of spatiotemporal wind characteristics on CNN model performance. The study demonstrates that regions with smaller deviations of autocorrelation coefficients are more favorable for CNN learning.
Article
Multidisciplinary Sciences
Tiankuang Zhou, Wei Wu, Jinzhi Zhang, Shaoliang Yu, Lu Fang
Summary: We propose a spatiotemporal photonic computing architecture to achieve dynamic processing, matching highly parallel spatial computing with high-speed temporal computing. A unified training framework is devised to optimize the physical system and the network model. The proposed architecture paves the way for ultrafast advanced machine vision and will find applications in unmanned systems, autonomous driving, ultrafast science, etc.
Article
Computer Science, Information Systems
K. U. Jaseena, Binsu C. Kovoor
Summary: Weather forecasting is the practice of predicting the state of the atmosphere based on different weather parameters. Accurate weather forecasts are crucial in various fields. With the advancement of atmospheric observing systems and the increasing volume of weather data, deep learning techniques are being used to improve weather prediction. This paper provides a comprehensive review of weather forecasting approaches and discusses potential future research directions.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Mechanics
Sudeepta Mondal, Soumalya Sarkar
Summary: Data-driven prediction of spatiotemporal fields in fluid flow problems has become increasingly important. However, the accuracy of prevalent approaches is often affected by the scarcity of data, especially when generating high-fidelity data is expensive. This article proposes a novel multi-fidelity spatiotemporal modeling approach to reduce the overhead of high-fidelity simulations and improve the accuracy of predictions.
Article
Engineering, Electrical & Electronic
Abdullah S. Alharthi, Alexander J. Casson, Krikor B. Ozanyan
Summary: The study aims to investigate the nature of gait variability by identifying gait intervals responsible for variations, using deep learning methods for sensor fusion. The research demonstrates successful classifications of gait signatures, achieving high F1-scores with CNN, and reveals that cognitive load can lower the classification accuracy.
IEEE SENSORS JOURNAL
(2021)
Article
Computer Science, Artificial Intelligence
D. Criado-Ramon, L. G. B. Ruiz, M. C. Pegalajar
Summary: This paper addresses the problem of electric demand prediction using neural networks and symbolization techniques. Symbolization techniques provide a shorter symbolic representation of a time series compared to the original time series. In the experimentation, the symbolization methodology resulted in a model that was trained significantly faster but had slightly worse quality metrics compared to the best numerical model.
APPLIED SOFT COMPUTING
(2022)
Article
Environmental Sciences
Andrew Bennett, Bart Nijssen
Summary: By embedding DL methods into PBHM models, it is possible to improve the accuracy of modeling hydrologic processes and achieve better predictive results.
WATER RESOURCES RESEARCH
(2021)
Article
Engineering, Civil
Pei Li, Mohamed Abdel-Aty, Shile Zhang
Summary: This study aims to improve the spatiotemporal transferability of a deep-learning crash likelihood prediction model by using transfer-learning approaches. The results suggest that the model can be accurately transferred to new data by using the fine-tuning transfer-learning approach, with higher predictive accuracy than models directly developed on the new data.
TRANSPORTATION RESEARCH RECORD
(2022)
Article
Ophthalmology
Lester C. Loschky, Ryan V. Ringer, Katrina Ellis, Bruce C. Hansen
Article
Engineering, Electrical & Electronic
Tyler E. Freeman, Lester C. Loschky, Bruce C. Hansen
SIGNAL PROCESSING-IMAGE COMMUNICATION
(2015)
Article
Clinical Neurology
Bruce C. Hansen, Andrew M. Haun, Aaron P. Johnson, Dave Ellemberg
Article
Neurosciences
Pavan Ramkumar, Bruce C. Hansen, Sebastian Pannasch, Lester C. Loschky
Article
Neurosciences
Reza Farivar, Simon Clavagnier, Bruce C. Hansen, Ben Thompson, Robert F. Hess
JOURNAL OF PHYSIOLOGY-LONDON
(2017)
Article
Ophthalmology
Bruno Richard, Bruce C. Hansen, Aaron P. Johnson, Patrick Shafto
Article
Neurosciences
Bruce C. Hansen, David J. Field, Michelle R. Greene, Cassady Olson, Vladimir Miskovic
Article
Neurosciences
Michelle R. Greene, Bruce C. Hansen
JOURNAL OF NEUROSCIENCE
(2020)
Article
Biochemical Research Methods
Bruce C. Hansen, Michelle R. Greene, David J. Field
Summary: Neuroimaging techniques such as fMRI and EEG have been used to study how visual information is transformed along the visual pathway. By leveraging the high temporal resolution of EEG, a new encoding technique based on distribution of responses from real-world scenes was developed, showing nonuniform transformations of scenes over time. This mapping technique offers a potential avenue for future studies to explore dynamic processes influencing high-level representations of the visual world.
PLOS COMPUTATIONAL BIOLOGY
(2021)
Article
Psychology, Multidisciplinary
Bruno Richard, Aaron P. Johnson, Benjamin Thompson, Bruce C. Hansen
FRONTIERS IN PSYCHOLOGY
(2015)
Article
Psychology, Experimental
Bruce C. Hansen, Pamela J. Rakhshan, Arnold K. Ho, Sebastian Pannasch