Filters








37 Hits in 8.2 sec

A review of uncertainty quantification in deep learning: Techniques, applications and challenges

Moloud Abdar, Farhad Pourpanah, Sadiq Hussain, Dana Rezazadegan, Li Liu, Mohammad Ghavamzadeh, Paul Fieguth, Xiaochun Cao, Abbas Khosravi, U. Rajendra Acharya, Vladimir Makarenkov, Saeid Nahavandi
2021 Information Fusion  
This study reviews recent advances in UQ methods used in deep learning, investigates the application of these methods in reinforcement learning, and highlights fundamental research challenges and directions  ...  Bayesian approximation and ensemble learning techniques are two widely-used types of uncertainty quantification (UQ) methods.  ...  [310] presented an uncertainty-aware consistency regularization technique for cross-domain semantic segmentation.  ... 
doi:10.1016/j.inffus.2021.05.008 fatcat:yschhguyxbfntftj6jv4dgywxm

A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges [article]

Moloud Abdar, Farhad Pourpanah, Sadiq Hussain, Dana Rezazadegan, Li Liu, Mohammad Ghavamzadeh, Paul Fieguth, Xiaochun Cao, Abbas Khosravi, U Rajendra Acharya, Vladimir Makarenkov, Saeid Nahavandi
2021 arXiv   pre-print
This study reviews recent advances in UQ methods used in deep learning. Moreover, we also investigate the application of these methods in reinforcement learning (RL).  ...  Bayesian approximation and ensemble learning techniques are two most widely-used UQ methods in the literature.  ...  [513] introduced an uncertainty-aware domain alignment approach to tackle the domain shift issue in the cross-domain UDA (Unsupervised Domain Adaptation) task.  ... 
arXiv:2011.06225v4 fatcat:wwnl7duqwbcqbavat225jkns5u

Deep Reinforcement Learning [article]

Yuxi Li
2018 arXiv   pre-print
Then we discuss important mechanisms for RL, including attention and memory, unsupervised learning, hierarchical RL, multi-agent RL, relational RL, and learning to learn.  ...  We start with background of artificial intelligence, machine learning, deep learning, and reinforcement learning (RL), with resources.  ...  Lanctot et al. (2017) observe that independent RL, in which each agent learns by interacting with the environment, oblivious to other agents, can overfit the learned policies to other agents' policies  ... 
arXiv:1810.06339v1 fatcat:kp7atz5pdbeqta352e6b3nmuhy

Data Science Competence Framework (Cf-Ds): Approach And Working Version

Yuri Demchenko, Adam Belloum, Tomasz Wiktorski
2016 Zenodo  
, and for companies to better define a set of required competences and skills for their specific industry domain in their search for Data Science talents, on the other hand.  ...  CF-DS will provide a basis for Data Science Body of Knowledge (DS-BoK) and Model Curriculum (MC-DC) definitions, and further for the Data Science Professional certification.  ...  stakeholders, primarily from research community and Research Infrastructures, but also involving industry via standardisation bodies, professional communities and directly via the project network.  ... 
doi:10.5281/zenodo.159438 fatcat:iq3ce3ygzbfvpmfgnxwyduappa

Edison Data Science Framework: Part 3. Data Science Model Curriculum (Mc-Ds) Release 2

Yuri Demchenko, Tomasz Wiktorski, Adam Belloum
2017 Zenodo  
When coupled with individual or group competence benchmarking, MC-DS can also be used for building individual training curricula and professional (self/up) skilling for effective career management.  ...  The MC-DS is built based on CF-DS and DS-BoK, where Learning Outcomes are defined based on CF-DS competences and Learning Units are mapped to Knowledge Units in DS-BoK.  ...  technology, algorithms, tools (including supervised, unsupervised, or reinforced learning) Level 3 SDSDA05 Apply Prescriptive Analytics methods Level 3 Cooperate with the CEN TC428 and  ... 
doi:10.5281/zenodo.1044357 fatcat:g5k7jybupfcexnd4a74vacanry

Edison Data Science Framework: Part 1. Data Science Competence Framework (Cf-Ds) Release 2

Yuri Demchenko, Adam Belloum, Tomasz Wiktorski
2017 Zenodo  
, and for companies to better define a set of required competences and skills for their specific industry domain in their search for Data Science talents, on the other hand.  ...  The EDISON project is designed to create a foundation for establishing a new profession of Data Scientist for European research and industry.  ...  and directly via the project network.  ... 
doi:10.5281/zenodo.1044345 fatcat:ybg432e6lfhgboplyo22ka37ci

Pedestrian Models for Autonomous Driving Part II: high level models of human behaviour [article]

Fanta Camara, Nicola Bellotto, Serhan Cosar, Florian Weber, Dimitris Nathanael, Matthias Althoff, Jingyuan Wu, Johannes Ruenz, André Dietrich, Gustav Markkula, Anna Schieben, Fabio Tango, Natasha Merat (+1 others)
2020 arXiv   pre-print
At these levels, early work has been done on descriptive and qualitative models of behaviour, but much work is still needed to translate them into quantitative algorithms for practical AV control.  ...  Autonomous vehicles (AVs) must share space with human pedestrians, both in on-road cases such as cars at pedestrian crossings and off-road cases such as delivery vehicles navigating through crowds on high-streets  ...  It is found that elderly people have a preference for road facilities that they consider to be safer such as pavements, pedestrian crossings, signalized intersections, cycle paths.  ... 
arXiv:2003.11959v1 fatcat:acjjwohahvdlxgy56j45fjtkdq

PUNCH4NFDI Consortium Proposal

The PUNCH4NFDI Consortium
2020 Zenodo  
that offers all storage and compute opportunities and all data transformation possibilities required for making the data fully productive in a sustainable way.  ...  Organised in 7 task areas, the consortium ultimately aims at establishing FAIR digital research products for its communities and beyond, spending their entire lifecycle inside a "science data platform"  ...  Those tools often provide alternative functionality to existing domain-specific solutions, but are often not yet validated for use with domain-specific data.  ... 
doi:10.5281/zenodo.5722894 fatcat:mfcvk55kqvgsthkp6dnxqyiyve

Engineering, Technology & Applied Science Research (ETASR), Vol. 11, No. 2, pp. 6845-7068 [article]

Various
2021 Zenodo  
The table of contents for this issue can be found at: https://etasr.com/index.php/ETASR/issue/view/68. For more info visit the journal's site at: www.etasr.com.  ...  Engineering, Technology & Applied Science Research (ETASR) is an international bimonthly wide scope, peer-reviewed open access journal for the publication of original articles concerned with diverse aspects  ...  Afef Hammami, director of the AFA center, and all its members for their cooperation in creating the AD dataset from Arabic voices.  ... 
doi:10.5281/zenodo.4720665 fatcat:mk2prflstjaa3bhkjenwy22s6u

25th Annual Computational Neuroscience Meeting: CNS-2016

Tatyana O. Sharpee, Alain Destexhe, Mitsuo Kawato, Vladislav Sekulić, Frances K. Skinner, Daniel K. Wójcik, Chaitanya Chintaluri, Dorottya Cserpán, Zoltán Somogyvári, Jae Kyoung Kim, Zachary P. Kilpatrick, Matthew R. Bennett (+597 others)
2016 BMC Neuroscience  
I will discuss theoretical results that point to functional advantages of splitting neural populations into subtypes, both in feedforward and recurrent networks.  ...  BMC Neuroscience 2016, 17(Suppl 1):A1 Neural circuits are notorious for the complexity of their organization.  ...  Allen and Jody Allen, for their vision, encouragement and support.  ... 
doi:10.1186/s12868-016-0283-6 pmid:27534393 pmcid:PMC5001212 fatcat:bt45etzj2bbolfcxlxo7hlv6ju

a need to know

William A. Spencer, Larry R. Gettman, H. H. Bridge, Allan J. Ryan
1983 Physician and sportsmedicine  
Using Unsupervised Learning Methods for Matching Nonprobability Samples to Probability Sample- FTrent Buskirk, Marketing Systems Group; David Dutwin, SSRS 11:55 a.m.  ...  Learning Network Dynamics via Regularized Tensor Decomposition-FYun-Jhong Wu, University of Michigan; Elizaveta Levina, University of Michigan; Ji Zhu, University of Michigan 9:35 a.m.  ...  Help SOI celebrate its centennial by visiting www.irs.gov/taxstats, learning more about the U.S. tax system, and supporting SOI's work as it prepares for the next 100 years. 192 Allen, Genevera 88, 93  ... 
doi:10.1080/00913847.1983.11708423 pmid:27409254 fatcat:7rlbreoerngb3epmighz7l5vr4

An Introduction to Probabilistic Programming [article]

Jan-Willem van de Meent, Brooks Paige, Hongseok Yang, Frank Wood
2021 arXiv   pre-print
We then discuss gradient-based maximum likelihood estimation in programs that are parameterized using neural networks, how to amortize inference using by learning neural approximations to the program posterior  ...  Foundational algorithms for this kind of language are discussed in the context of an interface between program executions and an inference controller.  ...  Most critically, while all of the authors were at Oxford together, three of them were explicitly supported at various times by the DARPA under its Probabilistic Programming for Advanced Machine Learning  ... 
arXiv:1809.10756v2 fatcat:cdcsscxbu5af7fpm5w6mjjiyra

Welcome Messages

2019 2019 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS)  
The database would be very useful to carry out deep learning research and various image processing tasks.  ...  The database along with its source code will be made open source for the research and academic purpose.  ...  CROSS CONDITIONAL NETWORK FOR SPEECH ENHANCEMENT We show the proposed network architecture in the Fig. 1 . The generator consists of two symmetric autoencoders.  ... 
doi:10.1109/ispacs48206.2019.8986291 fatcat:gu4zaxsqkncp5n2ebj5fybk7ce

Modeling the 3D Milky Way using Machine Learning with Gaia and infrared surveys [article]

David Cornu
2020 arXiv   pre-print
In a second part, I will propose a new method for reconstructing the 3D extinction distribution of the MW based on Convolutional Neural Networks (CNN).  ...  Concurrently, the past two decades have seen an explosion of the use of Machine Learning (ML) methods that are also increasingly employed in astronomy.  ...  There are obviously plenty of other frameworks that we did not describe here like Caffe, Theano, MXNET, scikit-learn (very suitable for teaching), etc.  ... 
arXiv:2010.01431v2 fatcat:pd7w4q3nijdp5mrxozon2z3mmq

Patient-generated evidence in Epidermolysis Bullosa (EB): Development of a questionnaire to assess the Quality of Life [chapter]

Laura Benedan, May El Hachem, Carlotta Galeone, Paolo Mariani, Cinzia Pilo, Gianluca Tadini
2021 Proceedings e report  
As a result of this disorder, patients face an unbearable burden in their lives, and their Quality of Life (QoL) is negatively affected at every life cycle stage.  ...  The methodological process to develop the questionnaire consisted of two phases: firstly, a critical review of scientific literature was performed; secondly, a pseudo-Delphi study was carried out.  ...  We acknowledge anonymous reviewer for the fruitful comments.  ... 
doi:10.36253/978-88-5518-461-8.38 fatcat:2yp72egbhrcvxng7avqoskbpse
« Previous Showing results 1 — 15 out of 37 results