136,337 Hits in 5.4 sec

Deep Residual Flow for Out of Distribution Detection [article]

Ev Zisselman, Aviv Tamar
2020 arXiv   pre-print
For out of distribution detection in image datasets, our approach provides a principled improvement over the state-of-the-art.  ...  We introduce the residual flow, a novel flow architecture that learns the residual distribution from a base Gaussian distribution.  ...  On deep networks trained for image classification, we obtain state-ofthe-art out-of-distribution detection performance.  ... 
arXiv:2001.05419v3 fatcat:niealvyeerh7ti5v7367pswnxq

Deep Residual Flow for Out of Distribution Detection

Ev Zisselman, Aviv Tamar
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
For out of distribution detection in image datasets, our approach provides a principled improvement over the state-of-the-art.  ...  We introduce the residual flow, a novel flow architecture that learns the residual distribution from a base Gaussian distribution.  ...  Acknowledgments This work is partly funded by the Israel Science Foundation (ISF-759/19) and the Open Philanthropy Project Fund, an advised fund of Silicon Valley Community Foundation.  ... 
doi:10.1109/cvpr42600.2020.01401 dblp:conf/cvpr/ZisselmanT20 fatcat:xzh55zfzfrdp5abeqdhflykwsy

Hybrid Models for Open Set Recognition [article]

Hongjie Zhang, Ang Li, Jie Guo, Yanwen Guo
2020 arXiv   pre-print
Existing methods fit a probability distribution to the training samples on their embedding space and detect outliers according to this distribution.  ...  Open set recognition requires a classifier to detect samples not belonging to any of the classes in its training set.  ...  This research was supported by the National Science Foundation of China under Grants 61772257 and the Fundamental Research Funds for the Central Universities 020914380080.  ... 
arXiv:2003.12506v2 fatcat:qilyie7ctfbzrag5j564un4i4u

Leak Localization in Water Distribution Networks using Deep Learning

Mohammadreza Javadiha, Joaquim Blesa, Adria Soldevila, Vicenc Puig
2019 2019 6th International Conference on Control, Decision and Information Technologies (CoDIT)  
This paper explores the use of deep learning for leak localization in Water Distribution Networks (WDNs) using pressure measurements.  ...  An innovative approach is proposed to convert every pressure residuals map to an image in order to apply a CNN.  ...  The leak detection is out of the scope of this work since it will be assumed that the leak detection is performed by an efficient method (as the night flow analysis [1] ).  ... 
doi:10.1109/codit.2019.8820627 dblp:conf/codit/JavadihaBSP19 fatcat:v5frznncjjgzhev4muxgqn5rfm

Label‐Free Leukemia Monitoring by Computer Vision

Minh Doan, Marian Case, Dino Masic, Holger Hennig, Claire McQuin, Juan Caicedo, Shantanu Singh, Allen Goodman, Olaf Wolkenhauer, Huw D. Summers, David Jamieson, Frederik V. Delft (+4 others)
2020 Cytometry Part A  
Cytometry Part A published by Wiley Periodicals, Inc. on behalf of International Society for Advancement of Cytometry.  ...  Adaptation to other types of leukemia is feasible, which would revolutionize residual disease monitoring. © 2020 The Authors.  ...  Fundamental to residual disease detection by flow cytometry is the characterization of a leukemia-associated immunophenotype at diagnosis.  ... 
doi:10.1002/cyto.a.23987 pmid:32091180 fatcat:7jbcwixnjrhhrctj2zpnkdxd5y

Distribution of injected pesticides in date palm trees

Adnan Al Samarrie, Abula Akela
2011 Agriculture and Biology Journal of North America  
The residues of four insecticides detected in dates 100 DPI, the insecticide residue levels of karate in the dates were 0.0034 & 0.019 of mixed and individual injected pesticides respectively which were  ...  It found that the systemic and non-systemic insecticides moved out of injection pores.  ...  rate was adjusted to prevent over flow out of the injection pore.  ... 
doi:10.5251/abjna.2011.2.12.1416.1426 fatcat:jep7d7wmbvcw5hhrqnnctehnye

Modeling the Hydrological Effect on Local Gravity at Moxa, Germany

Shaakeel Hasan, Peter A. Troch, J. Boll, C. Kroner
2006 Journal of Hydrometeorology  
To further explore the spatiotemporal dynamics of the relevant hydrological processes and their relation to observed gravity residuals, a GIS-based distributed hydrological model is applied for the Silberleite  ...  In this study time series analysis and distributed hydrological modeling techniques are applied to understand the effect of these hydrological processes on observed gravity residuals.  ...  We thank Wernfrid Kühnel, Matthias Meininger, Hennie Gertsen, and Roel Dijksma for their assistance in the field work and Henny van Lanen for providing his expertise on the issue of groundwater systems  ... 
doi:10.1175/jhm488.1 fatcat:k3dzxdgjn5fuxlqwjdqovc4dfm

Periodic Behavior of Deep Sea Current in the Hatoma Knoll Hydrothermal System [chapter]

Yasuo Furushima, Hiroyuki Yamamoto
2014 Subseafloor Biosphere Linked to Hydrothermal Systems  
The causes of the measurement result of the flow which appeared into a pulse form may indicate fluctuation of the blowout of hydrothermal water.  ...  Residual currents, which play an important role in material transport, tended to northwest direction and suggested that most of material might keep retaining within the caldera.  ...  Acknowledgements We thank the captain and crew of R/V Natsushima and the deep-sea submersible operation teams of HYPER-DOLPHIN 3000 for supporting the observation and for the installation and recovery  ... 
doi:10.1007/978-4-431-54865-2_50 fatcat:dcrypqvt6jczrjwr4wgt47cmti

Efficient Detection of Link-Flooding Attacks with Deep Learning

Chih-Hsiang Hsieh, Wei-Kuan Wang, Cheng-Xun Wang, Shi-Chun Tsai, Yi-Bing Lin
2021 Sustainability  
The DDoS attack is one of the most notorious attacks, and the severe impact of the DDoS attack on GitHub in 2018 raises the importance of designing effective defense methods for detecting this type of  ...  With the flexibility of software-defined networking, we design a novel framework and implement our ideas with a deep learning approach to improve the performance of the previous work.  ...  The link-flooding attack is a kind of distributed denial of service (DDoS) attack that does not directly attack the target devices but congests the target link with a large number of low-speed flows to  ... 
doi:10.3390/su132212514 fatcat:noymyaa4nnagpd6wdizmi3j3rm

A Wood Quality Defect Detection System Based on Deep Learning and Multicriterion Framework

Pingan Sun, C. Venkatesan
2022 Journal of Sensors  
The model introduced the deep learning mechanism and realized real-time and efficient reconstruction of multidimensional defect images of different wood by using the deep residual network for iterative  ...  It is proved that the system designed by the author can realize the timely detection of wood quality defects very effectively and save a lot of manpower and material resources.  ...  are input into the deep residual network based on a convolutional autoencoder for training, which can learn the data distribution characteristics of normal wood, but not the data distribution characteristics  ... 
doi:10.1155/2022/3234148 fatcat:qwowtjdxqjdytp3unpesa3dck4

Discrete Residual Flow for Probabilistic Pedestrian Behavior Prediction [article]

Ajay Jain, Sergio Casas, Renjie Liao, Yuwen Xiong, Song Feng, Sean Segal, Raquel Urtasun
2019 arXiv   pre-print
In this work, we propose the discrete residual flow network (DRF-Net), a convolutional neural network for human motion prediction that captures the uncertainty inherent in long-range motion forecasting  ...  In particular, our learned network effectively captures multimodal posteriors over future human motion by predicting and updating a discretized distribution over spatial locations.  ...  Acknowledgments We would like to thank Abbas Sadat for useful discussions during the development of this research.  ... 
arXiv:1910.08041v1 fatcat:mitmm4irprhjjbm656ed5gdhju

Hybrid deep fault detection and isolation: Combining deep neural networks and system performance models [article]

Manuel Arias Chao, Chetan Kulkarni, Kai Goebel, Olga Fink
2019 arXiv   pre-print
It provides superior results both for fault detection as well as for fault isolation.  ...  With the increased availability of condition monitoring data and the increased complexity of explicit system physics-based models, the application of data-driven approaches for fault detection and isolation  ...  W25 Flow into HPC lbm/s 24 W31 HPT coolant bleed lbm/s 25 W32 HPT coolant bleed lbm/s 26 W48 Flow out of HPT lbm/s 27 W50 Flow out of LPT lbm/s 28 epr Engine pressure ratio (P50/P2) - 29 SmFan Fan stall  ... 
arXiv:1908.01529v2 fatcat:ftmxwwawcvdepd77onfhqazkke

A Residual Learning-Based Network Intrusion Detection System

Jiarui Man, Guozi Sun, Weizhi Meng
2021 Security and Communication Networks  
In this paper, we propose a network intrusion detection model based on residual learning.  ...  Neural networks have been proved to perform well in network intrusion detection. In order to acquire better features of network traffic, more learning layers are necessarily required.  ...  after redundant flows and features are processed. e distribution is shown in Table 1 .  ... 
doi:10.1155/2021/5593435 fatcat:zy3oqpwbjzd6fbugwm6nvr3mcu

A Denoising Based Autoassociative Model for Robust Sensor Monitoring in Nuclear Power Plants

Ahmad Shaheryar, Xu-Cheng Yin, Hong-Wei Hao, Hazrat Ali, Khalid Iqbal
2016 Science and Technology of Nuclear Installations  
Sensors health monitoring is essentially important for reliable functioning of safety-critical chemical and nuclear power plants.  ...  Accuracy, autosensitivity, spillover, and sequential probability ratio test (SPRT) based fault detectability metrics are used for performance assessment and comparison with extensively reported five-layer  ...  Chad Painter (Director of Nuclear Power Plant Simulator Development and Training Program at International Atomic Energy Agency) for providing necessary tools and data to conduct this research.  ... 
doi:10.1155/2016/9746948 fatcat:mycxaxgjuzbxriodrcyhloq2lu

Why Normalizing Flows Fail to Detect Out-of-Distribution Data [article]

Polina Kirichenko, Pavel Izmailov, Andrew Gordon Wilson
2020 arXiv   pre-print
Detecting out-of-distribution (OOD) data is crucial for robust machine learning systems.  ...  Normalizing flows are flexible deep generative models that often surprisingly fail to distinguish between in- and out-of-distribution data: a flow trained on pictures of clothing assigns higher likelihood  ...  In this section, we test out-of-distribution detection using image representations from a deep neural network.  ... 
arXiv:2006.08545v1 fatcat:7etzvijmwffjpjpgf3h2bdbb7y
« Previous Showing results 1 — 15 out of 136,337 results