Filters








1,262 Hits in 5.6 sec

A Survey on Evidential Deep Learning For Single-Pass Uncertainty Estimation [article]

Dennis Ulmer
2021 arXiv   pre-print
Furthermore, they allow uncertainty estimation in a single model and forward pass by parameterizing distributions over distributions.  ...  This survey aims to familiarize the reader with an alternative class of models based on the concept of Evidential Deep Learning: For unfamiliar data, they admit "what they don't know" and fall back onto  ...  This allows to compute uncertainty in a single forward pass and set of weights.  ... 
arXiv:2110.03051v2 fatcat:nau5qmqvwrebfgmzk4edkzpbbe

A Survey of Uncertainty in Deep Neural Networks [article]

Jakob Gawlikowski, Cedrique Rovile Njieutcheu Tassi, Mohsin Ali, Jongseok Lee, Matthias Humt, Jianxiang Feng, Anna Kruspe, Rudolph Triebel, Peter Jung, Ribana Roscher, Muhammad Shahzad, Wen Yang (+2 others)
2022 arXiv   pre-print
Many researchers have been working on understanding and quantifying uncertainty in a neural network's prediction.  ...  For a practical application, we discuss different measures of uncertainty, approaches for the calibration of neural networks and give an overview of existing baselines and implementations.  ...  Khoshgoftaar, “A survey on image data augmen- in-domain uncertainty estimation and ensembling in deep learning,” in tation for deep learning,” Journal of Big Data, vol. 6, no. 1, pp.  ... 
arXiv:2107.03342v3 fatcat:cex5j3xq5fdijjdtdbt2ixralm

Evaluating Crowd Density Estimators via Their Uncertainty Bounds [article]

Jennifer Vandoni, Emanuel Aldea, Sylvie Le Hégarat-Mascle
2019 arXiv   pre-print
In this work, we use the Belief Function Theory which extends the probabilistic framework in order to provide uncertainty bounds to different categories of crowd density estimators.  ...  Our method allows us to compare the multi-scale performance of the estimators, and also to characterize their reliability for crowd monitoring applications requiring varying degrees of prudence.  ...  However, deep learning advancements significantly improved the state-of-the-art performance (see [1] for a comprehensive survey).  ... 
arXiv:1902.02831v1 fatcat:pd33k55aincntc6cb2xmxtyedu

Evaluating Crowd Density Estimators Via Their Uncertainty Bounds

Jennifer Vandoni, Emanuel Aldea, Sylvie Le Hegarat-Mascle
2019 2019 IEEE International Conference on Image Processing (ICIP)  
In this work, we use the Belief Function Theory which extends the probabilistic framework in order to provide uncertainty bounds to different categories of crowd density estimators.  ...  Our method allows us to compare the multi-scale performance of the estimators, and also to characterize their reliability for crowd monitoring applications requiring varying degrees of prudence.  ...  However, deep learning advancements significantly improved the state-of-the-art performance (see [1] for a comprehensive survey).  ... 
doi:10.1109/icip.2019.8803522 dblp:conf/icip/VandoniAH19 fatcat:m66hqt2oo5dl7ficcewvpoxjee

A Survey of Active Learning for Text Classification using Deep Neural Networks [article]

Christopher Schröder, Andreas Niekler
2020 arXiv   pre-print
We review AL for text classification using deep neural networks (DNNs) and elaborate on two main causes which used to hinder the adoption: (a) the inability of NNs to provide reliable uncertainty estimates  ...  For active learning (AL) purposes, NNs are, however, less commonly used -- despite their current popularity.  ...  Acknowledgements We thank Gerhard Heyer for his valuable feedback on the manuscript, Lydia Müller for fruitful discussions about the taxonomy and advice thereon, and Janos Borst for sharing his thoughts  ... 
arXiv:2008.07267v1 fatcat:joainuwblzbaplbls54tq4do3u

Handling of uncertainty in medical data using machine learning and probability theory techniques: A review of 30 years (1991-2020) [article]

Roohallah Alizadehsani, Mohamad Roshanzamir, Sadiq Hussain, Abbas Khosravi, Afsaneh Koohestani, Mohammad Hossein Zangooei, Moloud Abdar, Adham Beykikhoshk, Afshin Shoeibi, Assef Zare, Maryam Panahiazar, Saeid Nahavandi (+3 others)
2020 arXiv   pre-print
Nowadays, application of novel deep learning techniques to deal such uncertainties have significantly increased.  ...  Machine learning and probability theory methods have widespread application for this purpose in different fields.  ...  Funding The authors have not declared a specific grant for this research from any funding agency.  ... 
arXiv:2008.10114v1 fatcat:pgiep5djj5bdpe7qr4n2f4buky

From Shallow to Deep Interactions Between Knowledge Representation, Reasoning and Machine Learning (Kay R. Amel group) [article]

Zied Bouraoui and Antoine Cornuéjols and Thierry Denœux and Sébastien Destercke and Didier Dubois and Romain Guillaume and João Marques-Silva and Jérôme Mengin and Henri Prade and Steven Schockaert and Mathieu Serrurier and Christel Vrain
2019 arXiv   pre-print
This paper proposes a tentative and original survey of meeting points between Knowledge Representation and Reasoning (KRR) and Machine Learning (ML), two areas which have been developing quite separately  ...  , the semantic description of vector representations, the combination of deep learning with high level inference, knowledge graph completion, declarative frameworks for data mining, or preferences and  ...  Essentially, the RNN defined to represent a logic program P has all atoms of P in the input layer; one neuron, a kind of "and" gate, for each rule in a single hidden layer; and one neuron for every atom  ... 
arXiv:1912.06612v1 fatcat:yfnx3pzs6jhxtggaylc76pwjc4

Handling of uncertainty in medical data using machine learning and probability theory techniques: a review of 30 years (1991–2020)

Roohallah Alizadehsani, Mohamad Roshanzamir, Sadiq Hussain, Abbas Khosravi, Afsaneh Koohestani, Mohammad Hossein Zangooei, Moloud Abdar, Adham Beykikhoshk, Afshin Shoeibi, Assef Zare, Maryam Panahiazar, Saeid Nahavandi (+3 others)
2021 Annals of Operations Research  
Nowadays, various novel deep learning techniques have been proposed to deal with such uncertainties and improve the performance in decision making.  ...  Machine learning and probability theory methods have been widely used for this purpose in various fields.  ...  (Lim, 2020) 2020 COVID-19 Fallacies, facts and uncertainties about COVID-19 using Bayesian inference (Ghoshal et al., 2019) 2020 COVID-19 Uncertainty estimation in deep learning models for diagnosis  ... 
doi:10.1007/s10479-021-04006-2 pmid:33776178 pmcid:PMC7982279 fatcat:wq2eweu7hfduvbhwo5oaqaltoe

Uncertainty Aware AI ML: Why and How [article]

Lance Kaplan, Federico Cerutti, Murat Sensoy, Alun Preece, Paul Sullivan
2018 arXiv   pre-print
This paper argues the need for research to realize uncertainty-aware artificial intelligence and machine learning (AI\&ML) systems for decision support by describing a number of motivating scenarios.  ...  A theoretical demonstration illustrates how two emerging uncertainty-aware ML and AI technologies could be integrated and be of value for a route planning operation.  ...  Governments are authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation hereon.  ... 
arXiv:1809.07882v1 fatcat:uttcg75g6bhbpg3m5wsg7gfl3m

Tailored Uncertainty Estimation for Deep Learning Systems [article]

Joachim Sicking, Maram Akila, Jan David Schneider, Fabian Hüger, Peter Schlicht, Tim Wirtz, Stefan Wrobel
2022 arXiv   pre-print
Uncertainty estimation bears the potential to make deep learning (DL) systems more reliable.  ...  In this work, we propose a framework that, firstly, structures and shapes these requirements, secondly, guides the selection of a suitable uncertainty estimation method and, thirdly, provides strategies  ...  Poretschkin for fruitful discussions on the standardization and regulation of ML systems.  ... 
arXiv:2204.13963v1 fatcat:ofd4dibpxjemfeylt6hsc7puoa

Deep Learning Sensor Fusion for Autonomous Vehicle Perception and Localization: A Review

Jamil Fayyad, Mohammad A. Jaradat, Dominique Gruyer, Homayoun Najjaran
2020 Sensors  
Specifically, it focuses on recent studies that use deep learning sensor fusion algorithms for perception, localization, and mapping.  ...  , or manufacturing defects, among other factors; hence, it is not advisable to rely on a single sensor for any of the autonomous driving tasks.  ...  As an attempt to improve depth estimation, researchers in [82, 83] trained a deep learning CNN network to estimate depth using a single monocular camera.  ... 
doi:10.3390/s20154220 pmid:32751275 pmcid:PMC7436174 fatcat:fuhotalv2fdmbmpgx6llkp4xse

Scanning the Issue

Azim Eskandarian
2020 IEEE transactions on intelligent transportation systems (Print)  
The datasets repository is available at: https://sites.google.com/view/driveability-survey-dataset Deep Learning for Intelligent Transportation Systems: A Survey of Emerging Trends M. Veres and M.  ...  As the development of both robust methods and novel metrics depends on having access to large-scale driving datasets, a comprehensive and comparative study of 54 publicly available datasets for autonomous  ...  Temporal Signatures of Passive Wi-Fi Data for Estimating Bus Passenger Waiting Time at a Single Bus Stop P. Wepulanon, A. Sumalee, and W. H. K.  ... 
doi:10.1109/tits.2020.3008809 fatcat:etol5qoilvdnbj6gtjxk3gheaa

Unruptured Intracranial Aneurysms

J. Raymond, T. Nguyen, M. Chagnon, G. Gevry
2007 Interventional Neuroradiology  
Expert opinions are compatible with the primary hypothesis of a recently initiated randomized trial on unruptured aneurysms (TEAM), which is a benefit of endovascular treatment of 4% compared to observation  ...  Participants were more confident in their evaluation of treatment risks and in their skills at treating aneurysms than in their estimates of risks of rupture entailed by the presence of the lesion, the  ...  Acknowledgement The authors gratefully acknowledge the participation of the ABC WIN meeting attendees who took part in the survey; without your help and your patience this would not have been possible.  ... 
doi:10.1177/159101990701300302 pmid:20566114 pmcid:PMC3345486 fatcat:poqnxomxbfg4bhayc7in74eoeu

Multimodal Deep Generative Models for Trajectory Prediction: A Conditional Variational Autoencoder Approach [article]

Boris Ivanovic, Karen Leung, Edward Schmerling, Marco Pavone
2020 arXiv   pre-print
In this work, we provide a self-contained tutorial on a conditional variational autoencoder (CVAE) approach to human behavior prediction which, at its core, can produce a multimodal probability distribution  ...  yet easily accessible description of a data-driven, CVAE-based approach, highlight important design characteristics that make this an attractive model to use in the context of model-based planning for  ...  [1] for an extensive survey).  ... 
arXiv:2008.03880v2 fatcat:bwjfrdbxabhidbzwgbt5jmkagq

An Uncertainty-Informed Framework for Trustworthy Fault Diagnosis in Safety-Critical Applications [article]

Taotao Zhou, Enrique Lopez Droguett, Ali Mosleh, Felix T.S. Chan
2021 arXiv   pre-print
There has been a growing interest in deep learning-based prognostic and health management (PHM) for building end-to-end maintenance decision support systems, especially due to the rapid development of  ...  The fault diagnosis model flags the OOD dataset with large predictive uncertainty for expert intervention and is confident in providing predictions for the data within tolerable uncertainty.  ...  Section 2 summarizes the background of OOD detection, Bayesian deep learning, and a survey on its applications in PHM.  ... 
arXiv:2111.00874v1 fatcat:dyff3p67bjaabnkuq2eumnk2mi
« Previous Showing results 1 — 15 out of 1,262 results