Neural network google scholar. The semi-classical technique in field theory : some applic...

Neural network google scholar. The semi-classical technique in field theory : some applications (PhD thesis). Mar 16, 2026 · We show that deep neural networks can achieve dimension-independent rates of convergence for learning structured densities typical of image, audio, video, and text data. [8] He is University Professor Emeritus at the University of Toronto. The model includes sub-models: strategic layout model (SLM) for the layout phase and hierarchical battle model (HBM) for the battle phase. Hritwik Ghosh, Irfan Sadiq Rahat, Md. In fact, artificial neural-network techniques combine naturally with others forming a set of computational procedures with a solid theoretical base, and with an unquestionable efficiency in the resolution of real Aug 25, 2025 · This course module teaches the basics of neural networks: the key components of neural network architectures (nodes, hidden layers, activation functions), how neural network inference is performed, how neural networks are trained using backpropagation, and how neural networks can be used for multi-class classification problems. International Journal of Computer Vision 113, 1 (2015), 54–66. The motivation of this study is to provide the knowledge and understanding about various aspects of CNN. Search and read the full text of patents from around the world with Google Patents, and find prior art in our index of non-patent literature. Although we focus on and report performance of these methods as applied to training large neural networks, the underlying algorithms are applicable to any gradient-based machine learning . Spiking deep convolutional neural networks for energy-efficient object recognition. Abstract Abstract Though graph neural networks (GNNs)-based fraud detectors have received remarkable success in identifying fraudulent activities, few of them pay equal attention to models' performance and explainability. These tasks include ^ Christopher Bishop publications indexed by Google Scholar ^ a b "Professor Christopher Bishop elected Fellow of the Royal Society of Edinburgh". In this paper, we attempt to achieve high performance for graph-based fraud detection while considering model explainability. While their expressiveness is the reason they succeed, it also causes them to learn uninterpretable solutions that could have counter-intuitive properties. We describe the inspiration for artificial neural networks and how the methods of deep learning are built. 12 hours ago · ADS Google Scholar FU L, YAN S. While This research addresses industrial-scale data science challenges in precision agriculture through the development of an affordable embedded system for real-time weed detection. This study introduces the HGNN-HIL framework for evaluating Graph Mar 16, 2026 · Abstract In this work, we address the test-time adaptation challenge in graph neural networks (GNNs), focusing on overcoming the limitations in flexibility and generalization inherent in existing methods. This chapter examines the history of artificial neural networks research through the present day. However, important unsupervised problems on graphs, such as graph clustering, have proved more resistant to advances in GNNs. When the topology changes, the information bus and transmission line are aggregated and updated through the message‐passing mechanism in the graph neural network, and we enhance the approach by introducing current saturation as an Autocompletion of sentences is a very essential aspect of an intelligent writing system, but most n-gram models and unidirectional recurrent neural networks are unable to capture long-range and bidirectional contextual dependencies, leading to loss of semantic coherence and contextual accuracy. ‪PhD Candidate, University of Florida‬ - ‪‪Cited by 201‬‬ - ‪Adaptive Controls‬ - ‪Nonlinear Controls‬ - ‪Deep Learning‬ Google Scholar Citations lets you track citations to your publications over time. Effective Control of Complex process in power generation industries is a challenge to instrumentation engineers. 6, no. Yongqiang Cao, Yang Chen, and Deepak Khosla. Dropout, the most successful technique for regularizing neural networks, does not work well with RNNs and LSTMs. Mar 31, 2021 · It then presents convolutional neural networks (CNNs) which the most utilized DL network type and describes the development of CNNs architectures together with their main features, e. Released in beta in November 2004, the Google Scholar index includes peer-reviewed online academic journals and books, conference papers, theses and dissertations, preprints, abstracts, technical reports, and other Google publishes hundreds of research papers each year. Some key enabler deep learning algorithms such as generative adversarial networks, convolutional neural networks, and model transfers have completely changed our perception of information processing. The objective is to generate binary change maps from bi-temporal pre- and post-event satellite images. Dates and citation counts are estimated and are determined automatically by a computer program. From 2013 to 2023, he divided his time working for Google Brain and ‪University of Pennsylvania‬ - ‪‪Cited by 1,211‬‬ ‪Google Research‬ - ‪‪Cited by 3,716‬‬ - ‪Machine Learning‬ - ‪Graph Neural Networks‬ - ‪Foundation Models‬ - ‪AI for Science‬ We would like to show you a description here but the site won’t allow us. 1, 2025. Computer Science, University of Toronto‬ - ‪‪Cited by 1,018,445‬‬ - ‪machine learning‬ - ‪psychology‬ - ‪artificial intelligence‬ - ‪cognitive science‬ - ‪computer science‬ 1 Introduction Recurrent neural networks, long short-term memory [12] and gated recurrent [7] neural networks in particular, have been firmly established as state of the art approaches in sequence modeling and transduction problems such as language modeling and machine translation [29, 2, 5]. Change detection in satellite imagery across diverse geographic regions remains a core challenge in remote sensing. By applying advanced deep learning optimization techniques to YOLOv5 neural networks deployed on edge computing platforms, we demonstrate how industrial machine vision can be democratized for smaller operations. In this paper neural network based data validation method is developed and 5 days ago · An integrated photonic deep neural network was trained end-to-end with on-chip gradient-descent backpropagation, and all linear and nonlinear computations were performed on a single photonic chip 4 days ago · Physics-Informed Neural Networks (PINNs) have recently emerged as a promising framework for solving inverse problems by embedding physical laws directly into the training process of neural networks. Our Jun 17, 2025 · This paper proposes a nonlinear temporal method based on temporal features of long short‐term memory (LSTM) recurrent neural networks, which improves the accuracy of the LSTM method in forecasting steel plant load by extracting multiple related sequence temporal information. 5 days ago · An integrated photonic deep neural network was trained end-to-end with on-chip gradient-descent backpropagation, and all linear and nonlinear computations were performed on a single photonic chip 4 days ago · Physics-Informed Neural Networks (PINNs) have recently emerged as a promising framework for solving inverse problems by embedding physical laws directly into the training process of neural networks. Our Throughput-Optimized FPGA Accelerator for Deep Convolutional Neural Networks Deep convolutional neural networks (CNNs) have gained great success in various computer vision applications. The increasing application of machine learning (ML) in power system research necessitates validation under realistic conditions, with real-time hardware-in-the-loop (HIL) simulations serving as a critical tool for integrating computational models with physical systems to enable comprehensive testing across diverse scenarios. So, a sensor validation and data reconciliation methods are adopted for monitoring the data and faulty data are replaced. These methods have dramatically improved the state-of-the-art in speech recognition, visual object recognition, object detection and many ot … ‪Department of Mathematics and Statistics, Universiti Putra Malaysia‬ - ‪‪Cited by 637‬‬ - ‪Artificial Intelligence‬ - ‪Data Mining‬ - ‪Neural Networks‬ - ‪Discrete Optimization‬ - ‪Intelligent Systems‬ Abstract Convolutional Neural Networks (ConvNets) are commonly developed at a fixed resource budget, and then scaled up for better accuracy if more resources are available. Although a step-by-step tutorial of how to develop artificial neural networks is not included, additional reading suggestions covering artificial neural network Jul 9, 2025 · We propose a microgrid transient stability analysis method based on a message‐passing graph neural network. Google Scholar provides a simple way to broadly search for scholarly literature. In this study, we investigate current deep learning methods for change detection and propose a novel lightweight approach based on Temporal Convolutional Networks (TCNs). ‪University of Washington‬ - ‪‪Cited by 11,764‬‬ - ‪Artificial Intelligence‬ - ‪Machine Learning‬ - ‪Computer Vision‬ We would like to show you a description here but the site won’t allow us. Prediction of passenger flow of transit buses over a period of time using artificial neural network, In: Proc 3rd Int Congr Inf Commun Technol; 2018. Mar 16, 2026 · However, computing natural gradients for PINNs is prohibitively computationally costly and memory-intensive for all but small neural network architectures. Google Scholar DENG Y, CHEN H, LIU H, et al. We define the activation function and its role in Neural networks have been adapted to leverage the structure and properties of graphs. AlexNet is a convolutional neural network architecture developed for image classification tasks, notably achieving prominence through its performance in the ImageNet Large Scale Visual Recognition Challenge (ILSVRC). Guopeng LI Postdoc of Transport & Intelligent Vehicles, Delft University of Technology We would like to show you a description here but the site won’t allow us. 2015. Citations may include links to full text content from PubMed Central and publisher web sites. We develop a randomized algorithm for natural gradient descent for PINNs that uses sketching to approximate the natural gradient descent direction. Graph Neural Networks (GNNs) have achieved state-of-the-art results on many graph analysis tasks such as node classification and link prediction. ‪UC San Diego‬ - ‪‪Cited by 1,405‬‬ - ‪Computational Neuroscience‬ - ‪Artificial Intelligence‬ - ‪Vision‬ - ‪Audition‬ - ‪Language‬ ‪Shanghai University‬ - ‪‪Cited by 3,498‬‬ - ‪steganography‬ - ‪steganalysis‬ - ‪digital watermarking‬ - ‪digital forensics‬ - ‪AI Safety & Security‬ ‪KU Leuven‬ - ‪‪Cited by 729‬‬ - ‪Machine learning‬ - ‪optimization‬ - ‪deep learning‬ - ‪systems and control‬ The journal Neural Networks provides a forum for developing and nurturing an international community of scholars and practitioners who are interested in all aspects of neural networks, including deep learning and related approaches to artificial intelligence and machine learning. In this paper, we introduce a new perspective on the Bayesian Physics-Informed Neural Network (BPINN) framework, extending classical PINNs by explicitly incorporating training data generation A neural network-based method for predicting the recovery factors of different blocks is proposed, and the effects of injection volume, timing, concentration, and rate on the recovery factors are analyzed. SPIE, 2023: 127994R. In this paper, we show how to correctly apply dropout to LSTMs, and show that it substantially reduces overfitting on a variety of tasks. Moreover, conventional DNNs are mostly monocular vision based, whereas the human brain relies mainly on binocular vision. Retrieved 8 September 2020. ‪Professor, Peking University‬ - ‪‪Cited by 2,165‬‬ - ‪Computational Mechanics and Applied Mathematics‬ ‪King Abdullah University of Science and Technology / The Swiss AI Lab, IDSIA / University of Lugano‬ - ‪‪Cited by 315,636‬‬ - ‪computer science‬ - ‪artificial intelligence‬ - ‪reinforcement learning‬ - ‪neural networks‬ - ‪physics‬ ‪Assistant Professor of Radiology, Massachusetts General Hospital, Harvard Medical School‬ - ‪‪Cited by 1,937‬‬ - ‪Medical Image Analysis‬ - ‪Machine Learning‬ - ‪Computer Vision‬ - ‪Radiology‬ Geoffrey Everest Hinton (born 6 December 1947) is a British-Canadian computer scientist, cognitive scientist, cognitive psychologist and Nobel Prize laureate known for his work on artificial neural networks, which earned him the title "the Godfather of AI". Google Scholar provides a simple way to broadly search for scholarly literature. Nov 3, 2021 · Section 6 is the complete overview of neural networks enhanced by optimization algorithms, Section 7 is an application on artificial neural network-based optimization algorithms, Section 8 covers artificial neural network training-based optimized parameters and finally, Section 9 presents the conclusions and future work. We would like to show you a description here but the site won’t allow us. pp. She has more than 14 years of experience in teaching undergraduate courses and has in-depth knowledge in Computer networks, Software Engineering, OOAD, Computer Architecture, Software Quality and Management and Data Structures. This paper presents a Bidirectional Long Short-Term Memory (BiLSTM) based model as an improvement of We would like to show you a description here but the site won’t allow us. Jisan Mashrafi, Mohammed Abdul Al Arafat Tanzin, Sachi Nandan Mohanty, Shashi Kant, "Advanced neural network architectures for tomato leaf disease diagnosis in precision agriculture", Discover Sustainability, vol. It overcomes the limitations of traditional machine learning approaches. Event logs are widely used to record the status of high-tech systems, making log anomaly detection important for monitoring those systems. </p> Dec 21, 2024 · This study employs four distinct neural network models to predict an individual's vascular age using photoplethysmography (PPG), a non‐invasive, cost‐effective, and reliable technique. For example, in images, where each pixel becomes independent of the rest of the image when conditioned on pixels at most t steps away, a simple L2 -minimizing neural network can attain a rate of n-1/ ( (t+1)2+4), where t is Feb 1, 2019 · We introduce physics-informed neural networks – neural networks that are trained to solve supervised learning tasks while respecting any given laws of… Change detection in satellite imagery across diverse geographic regions remains a core challenge in remote sensing. Most importantly coordination of sensor functioning should be monitored regularly. Jun 17, 2025 · This paper proposes a nonlinear temporal method based on temporal features of long short‐term memory (LSTM) recurrent neural networks, which improves the accuracy of the LSTM method in forecasting steel plant load by extracting multiple related sequence temporal information. Unfortunately, making predictions using a whole ensemble of models is cumbersome and may be too computationally expensive to allow deployment to a large number of users, especially if the individual models are large neural nets. A series of experiments based on the ImageNet images and pre-trained MobileNet, Convolutional Neural Network (CNN) model from TensorFlow is used to show how an adversarial attack based on Images can be curated to outwit the machine learning model. The validation results from a case study of a high arch dam show that, compared with models such as CNN, extreme learning machine (ELM) and back propagation (BP) neural network, the proposed model exhibits higher accuracy and stronger stability in concrete dam deformation prediction. ^ a b Bishop, Christopher Michael (1983). In this paper, we introduce a new perspective on the Bayesian Physics-Informed Neural Network (BPINN) framework, extending classical PINNs by explicitly incorporating training data generation 12 hours ago · ADS Google Scholar FU L, YAN S. In this paper, we systematically study model scaling and identify that carefully balancing network depth, width, and resolution can lead to better performance. We propose a graph-based method for unsupervised log anomaly detection, dubbed Logs2Graphs, which first converts event logs into attributed, directed, and weighted graphs, and then leverages graph neural networks to perform graph-level anomaly detection Nov 17, 2025 · To address this, we propose a two-stage model called JFA, which incorporates hierarchical neural networks and knowledge-guided techniques. May 28, 2019 · Convolutional Neural Networks (ConvNets) are commonly developed at a fixed resource budget, and then scaled up for better accuracy if more resources are available. 12 hours ago · To effectively capture the potential nonlinear, dynamic, and temporal dependencies between vector (or scalar) responses and functional time series, we propose three novel frameworks that integrate recurrent neural networks with functional data characteristics: the functional recurrent neural network (FRNN), the functional long short-term memory Deep neural networks are highly expressive models that have recently achieved state of the art performance on speech and visual recognition tasks. Jan 14, 2022 · In this chapter, we go through the fundamentals of artificial neural networks and deep learning methods. The Neural Information Processing Systems Foundation is a non-profit corporation whose purpose is to foster the exchange of research advances in Artificial Intelligence and Machine Learning, principally by hosting an annual interdisciplinary academic conference with the highest ethical standards for a diverse and inclusive community. g. The book's coverage is broad, from supervised learning (prediction) to unsupervised learning. Mintajur Rahman Emon, Md. We present a simple regularization technique for Recurrent Neural Networks (RNNs) with Long Short-Term Memory (LSTM) units. , starting with the AlexNet network and closing with the High-Resolution network (HR. University of Edinburgh School of Informatics. Google Scholar is a freely accessible web search engine that indexes the full text or metadata of scholarly literature across an array of publishing formats and disciplines. DNNs developed as smaller vision agent networks associated with fundamental and less intelligent visual activities, can be combined to simulate more Feb 1, 2019 · We introduce physics-informed neural networks – neural networks that are trained to solve supervised learning tasks while respecting any given laws of… We would like to show you a description here but the site won’t allow us. Based on this observation, we propose a new scaling method that We show that these same techniques dramatically accelerate the training of a more modestly- sized deep network for a commercial speech recognition service. Sep 2, 2021 · A Gentle Introduction to Graph Neural Networks Neural networks have been adapted to leverage the structure and properties of graphs. The trained model is subsequently used for prediction. PubMed® comprises more than 40 million citations for biomedical literature from MEDLINE, life science journals, and online books. Jan 1, 2018 · Convolutional Neural Network (CNN) is a deep learning approach that is widely used for solving complex problems. We explore the components needed for building a graph neural network - and motivate the design choices behind them. Hypergraph neural network for gait recognition based on event camera [C]//3rd International Conference on Advanced Algorithms and Signal Image Processing, June 30–July 2, 2023, Kuala Lumpur, Malaysia. Sep 18, 2018 · Deep learning uses multiple layers to represent the abstractions of data to build computational models. Publishing our work enables us to collaborate and share ideas with, as well as learn from, the broader scientific… Jan 14, 2022 · In this chapter, we go through the fundamentals of artificial neural networks and deep learning methods. Net). Search across a wide variety of disciplines and sources: articles, theses, books, abstracts and court opinions. Numerous efforts have since continued to push the boundaries of recurrent language models and encoder 3 days ago · Extensive experiments on both self-built and public data sets validate its superior robustness and cross-data set generalization compare with several mainstream convolutional neural network (CNN)/transformer-based methods, underscoring its strong potential for real-word deployment in automated bridge crack detection. ‪Emeritus Prof. When the topology changes, the information bus and transmission line are aggregated and updated through the message‐passing mechanism in the graph neural network, and we enhance the approach by introducing current saturation as an Dec 6, 2023 · Abstract Psychologically faithful deep neural networks (DNNs) could be constructed by training with psychophysics data. Neural Networks wel… View full aims & scope ‪SRM Institute of Science and Technology‬ - ‪‪Cited by 29‬‬ - ‪Systems and Control‬ - ‪Clifford Algebra‬ - ‪Optimal Control‬ May 28, 2015 · Deep learning allows computational models that are composed of multiple processing layers to learn representations of data with multiple levels of abstraction. Nov 14, 2022 · In this work, the application of a feed-forward artificial neural network (FFANN) in predicting the degree of polymerization (DP) and loss of life (LOL) in oil-submerged transformers by using the solid insulation evaluation method is presented. Nov 19, 2016 · The concept of neural networks germinated independently but over time new contexts and disciplines have arisen, covering wider objectives which naturally include neural networks. Digital Library Google Scholar [2] Apr 23, 2020 · This prognostic study evaluates the performance and generalizability of a deep neural network trained on data from a single institution for classification of colorectal polyps on histopathologic slide images. 2 days ago · The workflow proposed in this paper introduces a neural network–based quantitative method for shale TOC prediction from geophysical data, providing a viable approach for effective TOC prediction and offering valuable support for shale oil and gas exploration and development. Modeling and simulation using differential and difference equations; Intelligent algorithms for optimization and decision-making; Hybrid methods combining neural networks with physical models; Symmetric and asymmetric analysis in nonlinear dynamics; Applications in energy systems, robotics, and cyber–physical systems. 599–608. The many topics include neural networks, support vector machines, classification trees and boosting---the first comprehensive treatment of this topic in any book. 3 days ago · Google Scholar Rane P, Kumar A. The components of artificial neural network architectures and both unsupervised and supervised learning methods are discussed. Oct 11, 2022 · This study aimed to investigate the diagnostic performance of the histogram array and convolutional neural network (CNN) based on diffusion-weighted imaging (DWI) with multiple b-values under magnetic resonance imaging (MRI) to distinguish pancreatic ductal adenocarcinomas (PDACs) from solid pseudopapillary neoplasms (SPNs) and pancreatic neuroendocrine neoplasms (PNENs). rchiu fdyxxt xvbwnuo mzwbzvj hvceaj vsboniv foibl apfb egxgt fwxshbq