Filters








5,285 Hits in 4.4 sec

Adapting Bayes network structures to non-stationary domains

Søren Holbech Nielsen, Thomas D. Nielsen
2008 International Journal of Approximate Reasoning  
When an incremental structural learning method gradually modifies a Bayesian network (BN) structure to fit a sequential stream of observations, we call the process structural adaptation.  ...  Structural adaptation is useful when the learner is set to work in an unknown environment, where a BN is gradually being constructed as observations of the environment are made.  ...  To adapt to the change, an immediate approach could be to learn a new network from the last k cases.  ... 
doi:10.1016/j.ijar.2008.02.007 fatcat:2hjh7c4d75fslfjnhhdsp4ee2i

Meta-Reinforcement Learning by Tracking Task Non-stationarity [article]

Riccardo Poiani, Andrea Tirinzoni, Marcello Restelli
2021 arXiv   pre-print
Many real-world domains are subject to a structured non-stationarity which affects the agent's goals and the environmental dynamics.  ...  However, most of the existing meta-RL algorithms for non-stationary domains either make strong assumptions on the task generation process or require sampling from it at training time.  ...  Introduction The ability to generalize and quickly adapt to non-stationary environments, where the dynamics and rewards might change through time, is a key component towards building lifelong reinforcement  ... 
arXiv:2105.08834v1 fatcat:vblxg6i35fgmjl4tkyhdtmdjeu

Wind Turbine Drivetrain Expert Fault Detection System: Multivariate Empirical Mode Decomposition based Multi-sensor Fusion with Bayesian Learning Classification

R. Uma Maheswari, R. Umamaheswari
2020 Intelligent Automation and Soft Computing  
The time-frequency representation is found to be very useful to analyze the non-stationary signals.  ...  The network structure is learned by conditional independent tests that unwarp the casual structure.  ... 
doi:10.32604/iasc.2020.013924 fatcat:4dzn6xzsajb43dpwm5zb2bexsi

A Review of Processing Methods and Classification Algorithm for EEG Signal

Yu Xie, Stefan Oniga
2020 Carpathian Journal of Electronic and Computer Engineering  
Digital filters EEG signals are random, non-stationary, and non-linear.  ...  It is the most commonly used method for non-stationary data where AR arguments are provided to the model. The expense exists in building the AR parameter model capability [5] .  ... 
doi:10.2478/cjece-2020-0004 fatcat:3bzo4p3yqvhmxbo2mydkqddwya

Spatial modelling using a new class of nonstationary covariance functions

Christopher J. Paciorek, Mark J. Schervish
2006 Environmetrics  
We also suggest non-Bayesian approaches to nonstationary kriging.  ...  In simulations, the nonstationary GP model adapts to function heterogeneity, unlike the stationary models, and also outperforms the other nonstationary models.  ...  changes over the domain and therefore the degree to which the model adapts to heterogeneity in the unknown function.  ... 
doi:10.1002/env.785 pmid:18163157 pmcid:PMC2157553 fatcat:fqpz2555yvgvzhci27fbz4pt3y

Detection and Classification of R-Peak Using Naïve Bayes Classifier

S Celin, K Vasanth
2018 International Journal of Engineering & Technology  
The best accuracy value of 99.7% is produced by using the Bayes classifiers in this paper.  ...  Due to their massively parallel structure, the neural networks derive their power, and it has the ability to learn from the experience.  ...  In feature extraction [16] wavelet transforms is used as a tool for processing non stationary signals like ECG signals.  ... 
doi:10.14419/ijet.v7i3.27.17982 fatcat:ckwfd5br7bhv3ljza47t3j2fqa

6G: Connecting Everything by 1000 Times Price Reduction [article]

Shunqing Zhang, Chenlu Xiang, Shugong Xu
2020 arXiv   pre-print
The commercial deployment of 5G communication networks makes the industry and academia to seriously consider the possible solutions for the next generation.  ...  Although it is impossible to identify every detail of 6G during the current time frame, we believe this article will help to eliminate the technical uncertainties and aggregate the efforts towards key  ...  ACKNOWLEDGMENT The authors would like to thank Prof. Xin Wang from Fudan University, Prof.  ... 
arXiv:2004.00853v1 fatcat:xqklvbc7zbc4tgqc5utikh6wja

When Naïve Bayes Nearest Neighbours Meet Convolutional Neural Networks [article]

Ilja Kuzborskij, Fabio Maria Carlucci, Barbara Caputo
2015 arXiv   pre-print
We address simultaneously the second and third by proposing a scalable version of Naive Bayes Non-linear Learning (NBNL, [2]).  ...  Results obtained using pre-trained CNNs on standard scene and domain adaptation databases show the strength of our approach, opening a new season for NBNNs.  ...  Applied to the NBNN learning framework, it results in a scalable Naïve Bayes Non-linear Learning technique (sNBNL).  ... 
arXiv:1511.03853v2 fatcat:cyawb3zmhnd7biifmqvm5zi72a

6G: Connecting Everything by 1000 Times Price Reduction

Shunqing Zhang, Chenlu Xiang, Shugong Xu
2020 IEEE Open Journal of Vehicular Technology  
The commercial deployment of 5G communication networks makes the industry and academia to seriously consider the possible solutions for the next generation.  ...  Although it is impossible to identify every detail of 6G during the current time frame, we believe this article will help to eliminate the technical uncertainties and aggregate the efforts towards key  ...  ACKNOWLEDGMENT The authors would like to thank Prof. Xin Wang from Fudan University, Prof.  ... 
doi:10.1109/ojvt.2020.2980003 fatcat:ncjm2oejmbcljhz2jpeqhpxfpu

Scalable Bayesian Modeling, Monitoring, and Analysis of Dynamic Network Flow Data

Xi Chen, Kaoru Irie, David Banks, Robert Haslinger, Jewell Thomas, Mike West
2018 Journal of the American Statistical Association  
We first develop flexible state-space models for streaming count data, able to adaptively characterize and quantify network dynamics efficiently in real-time.  ...  We then use these models as emulators of more structured, time-varying gravity models that allow formal dissection of network dynamics.  ...  Support from the Nakajima Foundation (to Kaoru Irie) is gratefully acknowledged.  ... 
doi:10.1080/01621459.2017.1345742 fatcat:t2vrggkrnraktgjnkvi2tvcbne

When Naïve Bayes Nearest Neighbors Meet Convolutional Neural Networks

Ilja Kuzborskij, Fabio Maria Carlucci, Barbara Caputo
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
We address simultaneously the second and third by proposing a scalable version of Naive Bayes Non-linear Learning (NBNL, [7] ).  ...  Results obtained using pre-trained CNNs on standard scene and domain adaptation databases show the strength of our approach, opening a new season for NBNNs.  ...  Applied to the NBNN learning framework, it results in a scalable Naïve Bayes Non-linear Learning technique (sNBNL).  ... 
doi:10.1109/cvpr.2016.231 dblp:conf/cvpr/KuzborskijCC16 fatcat:gwyus4ii6nbqpj7fszchazrkka

Use of Recurrent and Convolutional Neural Networks for the Problem of Long Term Condition Prediction for Equipment of an Oil-and-gas Production Enterprise

Iakov S. KOROVIN
2017 DEStech Transactions on Engineering and Technology Research  
In comparison with feed-forward networks, recurrent neural networks are oriented to processing of sequences of vectors, not single vectors of parameters.  ...  Here, the output signal of the network is a vector of probabilities, which shows that the current condition of the controlled object belongs to one or another predefined class from the knowledge base.  ...  It is impossible to eliminate completely the noise component from the initial signals, because external noise is non-stationary and random.  ... 
doi:10.12783/dtetr/eeta2017/7713 fatcat:wnwcez373vdltchzval64vmuom

Acoustic signal based traffic density state estimation using adaptive Neuro-Fuzzy classifier

Prashant Borkar, L. G. Malik
2013 2013 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)  
Adaptive Neuro-Fuzzy classifier is used to model the traffic density state as Low (40 Km/h and above), Medium (20-40 Km/h), and Heavy (0-20 Km/h).  ...  The (Scaled Conjugate Gradient) SCG algorithm, which is a supervised learning algorithm for network-based methods, is used to computes the second-order information from the two first-order gradients of  ...  afford quasi stationary signal behavior. • Time-frequency domain feature generation consider the non-stationary signal behavior of passing vehicles and it lead to accurate measures of signal energies  ... 
doi:10.1109/fuzz-ieee.2013.6622444 dblp:conf/fuzzIEEE/BorkarM13 fatcat:ztv4yhop5bd2vpqgwgvfryeaxy

Bayesian networks in neuroscience: a survey

Concha Bielza, Pedro Larrañaga
2014 Frontiers in Computational Neuroscience  
They have been shown to be powerful tools to encode dependence relationships among the variables of a domain under uncertainty.  ...  In this paper we review Bayesian networks and how they can be learned automatically from data by means of structure learning algorithms.  ...  Higherorder and non-stationary Markov models allow more complex temporal processes. However, such complex models pose obvious challenges to structure and parameter estimation. Example.  ... 
doi:10.3389/fncom.2014.00131 pmid:25360109 pmcid:PMC4199264 fatcat:2ip7hztt4fexdj5cw4a2gpmgbu

Common Spatial Filter for Improving the Classification of EEG using Artificial Neural Network

Shreyas J, Bhavani D, Udayaprasad P K, Srinidi N N, Dharamendra Chouhan, S M Dilip Kumar
2021 Zenodo  
Common spatial pattern is used in feature extraction for the improvement of the classifier of different subjects and tested with artificial neural network (ANN).  ...  To overcome this issue, common spatial pattern (CSP) method is proposed. The dataset contains 9 subjects EEG data.  ...  To accomplish this, a two-stream structure is established, one in the source domain and another in the destination domain.  ... 
doi:10.5281/zenodo.5805957 fatcat:6fnpncbz25dcbolxufznkx6ghy
« Previous Showing results 1 — 15 out of 5,285 results