Filters








1,268 Hits in 3.8 sec

Rank Projection Trees for Multilevel Neural Network Interpretation [article]

Jonathan Warrell, Hussein Mohsen, Mark Gerstein
2018 arXiv   pre-print
A variety of methods have been proposed for interpreting nodes in deep neural networks, which typically involve scoring nodes at lower layers with respect to their effects on the output of higher-layer  ...  Here, we outline a flexible framework which may be used to generate multiscale network interpretations, using any previously defined scoring function.  ...  Figure 1 : Rank projection tree schematic. The rank projection tree (left, T ) is mapped onto a trained neural network (right, N ) via the mapping φ which depends on an arbitrary ranking function r.  ... 
arXiv:1812.00172v1 fatcat:ohk5wa7d5bdebavd4ptrzpcqlu

Face recognition with Multilevel B-Splines and Support Vector Machines

Manuele Bicego, Gianluca Iacono, Vittorio Murino
2003 Proceedings of the 2003 ACM SIGMM workshop on Biometrics methods and applications - WBMA '03  
The obtained results are very encouraging, outperforming traditional methods like eigenface, elastic matching or neural-networks based recognition systems.  ...  Such heightfields are approximated using Multilevel B-Splines, and the coefficients of approximation are used as features for the classification process, which is performed using Support Vector Machines  ...  Then, the feature vector is classified by a radial basis function neural network. Ayinde et al.  ... 
doi:10.1145/982507.982511 fatcat:w3ez5xdwbbc3jkajxfjdeuftfy

Intellectual Multi-Level System for Neuro-Fuzzy and Cognitive Analysis and Forecast of Scientific-Technological and Innovative Development

Sergey Gorbachev, E. Siemens, A.D. Mehtiyev, V.I. Syryamkin, A.V. Yurchenko
2018 MATEC Web of Conferences  
The article is devoted to description of structure and operation algorithms of a multilevel intellectual system for analysis and forecast of scientific-technological and innovative development of objects  ...  The authors have built a neutrosophic cognitive map for interaction of technologies based on the results of foresight studies.  ...  Rumyantseva from the National Research Tomsk State University for her assistance in preparing the paper.  ... 
doi:10.1051/matecconf/201815501012 fatcat:vypbmvet7vc4ffticwt2vza7ie

Contrasting Classical and Machine Learning Approaches in the Estimation of Value-Added Scores in Large-Scale Educational Data

Jessica Levy, Dominic Mussack, Martin Brunner, Ulrich Keller, Pedro Cardoso-Leite, Antoine Fischbach
2020 Frontiers in Psychology  
The different models include linear and non-linear methods and extend classical models with the most commonly used machine learning methods (i.e., random forest, neural networks, support vector machines  ...  To date, the two most common statistical models used for the calculation of VA scores are two classical methods: linear regression and multilevel models.  ...  More concretely, the multilevel model performed better than all the other models and the neural network worse.  ... 
doi:10.3389/fpsyg.2020.02190 pmid:32973639 pmcid:PMC7472739 fatcat:isw4bfx7eban5mq3muqr2ybnyy

An Approach Based on Multilevel Convolution for Sentence-Level Element Extraction of Legal Text

Zhe Chen, Hongli Zhang, Lin Ye, Shang Li, Peng Li
2021 Wireless Communications and Mobile Computing  
The encoder applies multilevel convolutional neural networks (CNN) and Long Short-Term Memory (LSTM) as feature extraction networks to extract local neighborhood and context information from legal text  ...  To our best knowledge, it is one of the first attempts to apply a multilabel classification algorithm for element extraction of legal text.  ...  tional neural network for improving short text classification,” [12] G. Kurata, B. Xiang, and B. Zhou, “Improved neural network- Neurocomputing, vol. 174, pp. 806–814, 2016.  ... 
doi:10.1155/2021/1043872 fatcat:tvqftw2d25dfxf6vt6v3sfsjf4

Global Context-Based Multilevel Feature Fusion Networks for Multilabel Remote Sensing Image Scene Classification

Xin Wang, Lin Duan, Chen Ning
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Experimental results demonstrate that the proposed method is superior to some popular networks for multilabel RS image scene classification.  ...  feature fusion network.  ...  Attention Mechanisms Embedding attention mechanisms in neural networks has been shown to be an effective way to improve the performance of the original networks [29] .  ... 
doi:10.1109/jstars.2021.3122464 fatcat:iumvszpu25dszdgp5fkpxqqudq

Explainable Artificial Intelligence (XAI) to Enhance Trust Management in Intrusion Detection Systems Using Decision Tree Model

Basim Mahbooba, Mohan Timilsina, Radhya Sahal, Martin Serrano, Ahmed Mostafa Khalil
2021 Complexity  
We use simple decision tree algorithms that can be easily read and even resemble a human approach to decision-making by splitting the choice into many small subchoices for IDS.  ...  The previous studies focused more on the accuracy of the various classification algorithms for trust in IDS.  ...  Nowadays, due to highly accurate predictions, the deep neural network (DNN) is getting more popular. ese kinds of models are useful, but they are hard to interpret.  ... 
doi:10.1155/2021/6634811 fatcat:lcvavyqwr5etxkzzg5aeqddyse

Software Fault Prediction with Data Mining Techniques by Using Feature Selection Based Models

Amit Kumar Jakhar, Department of Computer Science and Engineering, Birla Institute of Technology, Mesra, Ranchi, Jharkhand, India, Kumar Rajnish, Department of Computer Science and Engineering, Birla Institute of Technology, Mesra, Ranchi, Jharkhand, India
2018 International Journal on Electrical Engineering and Informatics  
This prediction helps project managers to effectively utilize the resources for better quality assurance.  ...  Therefore, it is important to find suitable and significant measures which are most relevant for finding the defects in the software system.  ...  Neural Networks (NN) Neural networks consists of numerous interconnected neurons (processing elements).  ... 
doi:10.15676/ijeei.2018.10.3.3 fatcat:zgjwprx6rjflll5yzy3twcvgty

Using machine learning as a surrogate model for agent-based simulations

Claudio Angione, Eric Silverman, Elisabeth Yaneske, Roland Bouffanais
2022 PLoS ONE  
Our results suggest that, in most scenarios, artificial neural networks (ANNs) and gradient-boosted trees outperform Gaussian process surrogates, currently the most commonly used method for the surrogate  ...  Here we compare multiple machine-learning methods for ABM surrogate modelling in order to determine the approaches best suited as a surrogate for modelling the complex behaviour of ABMs.  ...  Unlike GPs, methods like neural networks, non-linear SVM and gradient-boosted trees do not bring with them insightful uncertainty quantification measures 'for free'.  ... 
doi:10.1371/journal.pone.0263150 pmid:35143521 pmcid:PMC8830643 fatcat:6on2y5yq45dsdmx7ss5ivc64j4

A Hierarchical Urban Forest Index Using Street-Level Imagery and Deep Learning

Philip Stubbings, Joe Peskett, Francisco Rowe, Dani Arribas-Bel
2019 Remote Sensing  
First, areas of vegetation are detected within street-level imagery using a state-of-the-art deep neural network model.  ...  spaces where shallow trees are abundant, in high density residential areas with backyard trees, and along street networks with high density of high trees.  ...  PSPNet relies on a Fully Convolutional Neural Network (FCNN) for pixel prediction and a pyramid parsing module for harvesting sub-region image representations.  ... 
doi:10.3390/rs11121395 fatcat:nyph7ljy3zdhxemhz2tebuke7u

Survival prediction models since liver transplantation - comparisons between Cox models and machine learning techniques

Georgios Kantidakis, Hein Putter, Carlo Lancia, Jacob de Boer, Andries E. Braat, Marta Fiocco
2020 BMC Medical Research Methodology  
Neural networks show better performance than both Cox models and random survival forest based on the Integrated Brier Score at 10 years.  ...  Criticism to ML is related to unsuitable performance measures and lack of interpretability which is important for clinicians.  ...  Acknowledgements The authors would like to thank the United Network of Organ Sharing (UNOS) and Scientific Registry of Transplant Recipients (SRTR) for providing the data about liver transplantation to  ... 
doi:10.1186/s12874-020-01153-1 pmid:33198650 fatcat:cxxukllup5c4piybazto5t2w5u

2020 Index IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Vol. 13

2020 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
., +, JSTARS 2020 1729-1739 Multilevel Deep Learning Network for County-Level Corn Yield Estimation in the U.S. Corn Belt.  ...  ., +, JSTARS 2020 1119-1133 Hyperspectral Anomaly Detection Based on Low-Rank Representation With Data-Driven Projection and Dictionary Construction.  ...  A New Deep-Learning-Based Approach for Earthquake-Triggered Landslide Detection From Single-Temporal RapidEye Satellite Imagery. Yi, Y., +, JSTARS 2020  ... 
doi:10.1109/jstars.2021.3050695 fatcat:ycd5qt66xrgqfewcr6ygsqcl2y

Graph-Partitioning-Based Diffusion Convolutional Recurrent Neural Network for Large-Scale Traffic Forecasting [article]

Tanwi Mallick, Prasanna Balaprakash, Eric Rask, Jane Macfarlane
2020 arXiv   pre-print
We present an approach for implementing a DCRNN for a large highway network that overcomes these limitations.  ...  Recently, diffusion convolutional recurrent neural networks (DCRNNs) have achieved state-of-the-art results in traffic forecasting by capturing the spatiotemporal dynamics of the traffic.  ...  We develop an overlapping nodes approach, an improvement strategy for the graph-partitioning-based DCRNN that includes sensor locations from partitions that are geographically close to a given partition  ... 
arXiv:1909.11197v4 fatcat:ugwujpeh7rhzdptk3uztjdxopa

Fast content-based image retrieval using convolutional neural network and hash function

Domonkos Varga, Tamas Sziranyi
2016 2016 IEEE International Conference on Systems, Man, and Cybernetics (SMC)  
The success of deep learning techniques such as convolutional neural networks have motivated us to explore its applications in our context.  ...  We are very thankful to Levente Kovács for helping us with professional advices in high-performance computing.  ...  A ranking list was applied to encode the multilevel similarity information. Wang et al.  ... 
doi:10.1109/smc.2016.7844637 dblp:conf/smc/VargaS16 fatcat:vj5wgjkarzcxddfypouse3s6i4

A robust and interpretable end-to-end deep learning model for cytometry data

Zicheng Hu, Alice Tang, Jaiveer Singh, Sanchita Bhattacharya, Atul J. Butte
2020 Proceedings of the National Academy of Sciences of the United States of America  
In addition, we developed a permutation-based method for interpreting the deep convolutional neural network model.  ...  In this study, we propose and test a deep convolutional neural network for analyzing cytometry data in an end-to-end fashion, allowing a direct association between raw cytometry data and the clinical outcome  ...  We thank Patrick Dunn, Elizabeth Thomson, Henry Schaefer, Daniel Wong, Dmytro Lituiev, Benjamin Glicksberg, and Matthew Elliott for helpful discussions. We thank Boris Oskotsky for server support.  ... 
doi:10.1073/pnas.2003026117 pmid:32801215 fatcat:obx4nw3bsnb3zm6y4ppwtut4ei
« Previous Showing results 1 — 15 out of 1,268 results