Filters








1,122 Hits in 6.5 sec

Improving cross-validated bandwidth selection using subsampling-extrapolation techniques

Qing Wang, Bruce G. Lindsay
2015 Computational Statistics & Data Analysis  
h i g h l i g h t s • A two-stage subsampling-extrapolation bandwidth selection procedure is proposed. • An automatic nested cross-validation method is developed to select the subsample size. • The extrapolated  ...  Furthermore, we developed a second-order extrapolation technique as an extension designed to improve the approximation of the true optimal bandwidth.  ...  We want to thank the three anonymous reviewers for their valuable comments that lead to a much improved version of the manuscript.  ... 
doi:10.1016/j.csda.2015.03.005 fatcat:pjmbrnpcivgltox2b7fxhioivi

AN EFFICIENT HYBRID ALGORITHM TO REDUCE LATENCY IN AD-HOC AGGREGATION

Rajkumar
2014 Journal of Computer Science  
Peer to Peer networks are used for distribution and sharing of documents.  ...  In traditional techniques, when aggregate functions like average, sum and count are encountered, the aggregate operation is performed by considering all the nodes and tuples, which reduces the efficiency  ...  Hybrid algorithm is used which involves two phases with gossiping thereby selecting peers randomly and then random selection of tuples is done. Finally cross validation is performed.  ... 
doi:10.3844/jcssp.2014.1186.1196 fatcat:p5qzuoq5cnad3h2r6nvfa6m7dq

Semiparametric varying-coefficient partially linear models with auxiliary covariates

Xiaojing Wang, Yong Zhou, Yang Liu
2018 Statistics and its Interface  
In this paper, we consider a semiparametric varyingcoefficient partially linear model when some of the covariates are only measured on a selected validation set whereas auxiliary variables are observed  ...  We use a Boston Housing dataset to illustrate the performance of the proposed method in practice.  ...  A way using cross validation score function will be introduced here for selecting the optimal bandwidth to construct nonparametric estimators in Model (1.1).  ... 
doi:10.4310/sii.2018.v11.n4.a4 fatcat:zdsinnrp5zfwngcjci4xgxrw2m

Using machine learning to maintain pub/sub system QoS in dynamic environments

Joe Hoffert, Daniel Mack, Douglas Schmidt
2009 Proceedings of the 8th International Workshop on Adaptive and Reflective MIddleware - ARM '09  
These techniques include decision trees, neural networks, and linear logistic regression classifiers that can be trained on existing data to interpolate and extrapolate for new data.  ...  The results of our work thus far show that decision trees and neural networks can effectively classify the best protocols to use.  ...  All three techniques were trained using n-fold cross validation, where each fold partitions the data into a training set and testing set.  ... 
doi:10.1145/1658185.1658189 dblp:conf/middleware/HoffertMS09 fatcat:5kxhzvcyubgvzgewqm3nuvusha

Combining the regression discontinuity design and propensity score-based weighting to improve causal inference in program evaluation

Ariel Linden, John L. Adams
2012 Journal of Evaluation In Clinical Practice  
One clear advantage of the weighting technique over regression covariate adjustment is that we can directly inspect the degree to which balance was achieved.  ...  In this paper, we describe key advances in the RD design over the past decade and illustrate their implementation using data from a health management intervention.  ...  Researchers can choose between 'rule-of-thumb' estimators [16] , cross-validation procedures [7] or data-driven techniques [18] .  ... 
doi:10.1111/j.1365-2753.2011.01768.x pmid:22304484 fatcat:ukttbthg4rcmzoprfazzhnklnu

A Robust and Efficient Deep Learning Method for Dynamical Mass Measurements of Galaxy Clusters [article]

Matthew Ho, Markus Michael Rau, Michelle Ntampaka, Arya Farahi, Hy Trac, Barnabas Poczos
2019 arXiv   pre-print
We then evaluate the performance of each model on an independent set of mock observations selected from the same simulated catalog.  ...  We train each model as a regression over cluster mass using a labeled catalog of realistic mock cluster observations generated from the MultiDark simulation and UniverseMachine catalog.  ...  The CosmoSim database used in this paper is a service by the Leibniz-Institute for Astrophysics Potsdam (AIP).  ... 
arXiv:1902.05950v1 fatcat:4gp2u2kdijfovciiuxe6irly6i

An SVM-Based Classifier for Estimating the State of Various Rotating Components in Agro-Industrial Machinery with a Vibration Signal Acquired from a Single Point on the Machine Chassis

Ruben Ruiz-Gonzalez, Jaime Gomez-Gil, Francisco Gomez-Gil, Víctor Martínez-Martínez
2014 Sensors  
Secondly, the SVM-based system accuracy was evaluated by using Leave-One-Out cross-validation, with the selected features as the input data.  ...  an SVM classifier, an 85% mean cross-validation accuracy can be reached, which only requires a maximum of seven features as its input, and no significant improvements are noted between the use of either  ...  The k-fold cross-validation repeats this training and test process k times, using each of the k subsamples only once as the test dataset.  ... 
doi:10.3390/s141120713 pmid:25372618 pmcid:PMC4279508 fatcat:sv427dbylfdvxkhbvodboh5wg4

Indirect Off-Axis Holography for Antenna Metrology [chapter]

Ana Arboleya, Jaime Laviada, Juha Ala-Laurinaho, Yuri Álvarez, Fernando Las-Heras, Antti V. Räisänen
2017 Holographic Materials and Optical Systems  
This technique can be used to avoid the effect of the errors related to the phase acquisition and to further develop new efficient and robust techniques capable of phase retrieval from amplitude-only acquisitions  ...  Last sections are devoted to the description of novel measurement techniques developed by the authors in order to overcome the main limitations of the conventional methods.  ...  Thus, the quality of the phase retrieval is improved.  ... 
doi:10.5772/67294 fatcat:doac4n466rbijgfkrokyxxlpky

The temporal overfitting problem with applications in wind power curve modeling [article]

Abhinav Prakash, Rui Tuo, Yu Ding
2022 arXiv   pre-print
Using existing model selection methods, like cross validation, results in model overfitting in presence of temporal autocorrelation.  ...  Our approach yields significant improvement when predicting response for a time period different from the training time period.  ...  We employ a forward stepwise subset selection using a five-fold cross validation to get the best subset of input variables.  ... 
arXiv:2012.01349v3 fatcat:mwkssazatbfp7gagctklc5dmkq

An independently corroborated, diatom-inferred record of long-term drought cycles occurring over the last two millennia in New Mexico, USA
English

Jay Hodgson, Amelia Ward, Clifford Dahm
2013 INLAND WATERS  
We investigated late Holocene (2000 YBP to present) drought in northern New Mexico, USA, using diatom valves sampled from a lake sediment core.  ...  Also, both the tree ring and diatom records were tested for concordance with indirect radiocarbon solar intensity data and were both significantly cross-correlated with solar intensity.  ...  Davis for their tremendous efforts helping us with sediment coring. Discussions with R.S. Anderson and D. Davis greatly helped us select our study site. G.M. Ward, R. Findlay, and J.  ... 
doi:10.5268/iw-3.4.629 fatcat:wbnq2utr7bh37grbulv7s5uchq

Sparse Signal Models for Data Augmentation in Deep Learning ATR [article]

Tushar Agarwal, Nithin Sugavanam, Emre Ertin
2022 arXiv   pre-print
We propose a data augmentation approach to incorporate domain knowledge and improve the generalization power of a data-intensive learning algorithm, such as a Convolutional neural network (CNN).  ...  Using this estimated model, we synthesize new images at poses and sub-pixel translations not available in the given data to augment CNN's training data.  ...  We also use a validation data-set, D val = {(X u , Y u )} N val u=1 for cross-validation during training and a test data-set, D test = {(X u , Y u )} Ntest u=1 for evaluating g(X; w) post-training.  ... 
arXiv:2012.09284v2 fatcat:mebuuhqvqnhtrlxm2u2muqjyu4

Estimation of time-varying delay time in nonstationary linear systems: an approach to monitor human operator adaptation in manual tracking tasks

E.R. Boer, R.V. Kenyon
1998 IEEE transactions on systems, man and cybernetics. Part A. Systems and humans  
Our goal to obtain fractional delay time estimates was realized by using the bicubic interpolation scheme as part of the EKF to provide subsample magnitude and derivative estimates of the observed input  ...  Human operator models were refined as new identification techniques became available.  ...  Again common sense should be used in selecting these model coefficient estimates. G.  ... 
doi:10.1109/3468.650325 fatcat:l35lxdcxfnfbpm3jhcxgomvhky

Automated acoustic detection of mouse scratching

Peter Elliott, Max G'Sell, Lindsey M. Snyder, Sarah E. Ross, Valérie Ventura, Andrey E Ryabinin
2017 PLoS ONE  
This report is the first method to apply supervised learning techniques to automate acoustic scratch detection.  ...  In animal models of itch, scratching behavior is frequently used as a proxy for itch, and this behavior is typically assessed through visual quantification.  ...  The neighborhood adjustment method (Sec 3.2.2) gives an improvement over the unadjusted cross prediction. https://doi.org/10.1371/journal.pone.0179662.g005 Table 2 summarizes the behavioral sources of  ... 
doi:10.1371/journal.pone.0179662 pmid:28678797 pmcid:PMC5497976 fatcat:2p6adrbchbcm7kpiibtepxv2la

Scalable Remote Rendering with Depth and Motion-flow Augmented Streaming

Dawid Paja̧k, Robert Herzog, Elmar Eisemann, Karol Myszkowski, Hans-Peter Seidel
2011 Computer graphics forum (Print)  
Our framework is progressive, scalable, and allows us to stream augmented high-resolution (e.g., HDready) frames with small bandwidth on standard hardware.  ...  Examples include spatio-temporal upsampling (1, 2), 3D stereo rendering (3), or frame extrapolation (4).  ...  We use the technique proposed in [DRE * 10], where a grid is overlayed with the depth buffer.  ... 
doi:10.1111/j.1467-8659.2011.01871.x fatcat:utsg7hmcmrhz5oh2bgavgsqhcm

Reality Engine graphics

Kurt Akeley
1993 Proceedings of the 20th annual conference on Computer graphics and interactive techniques - SIGGRAPH '93  
While many team members made important contributions to the design, I especially acknowledge Mark Leather for developing the multisample antialiasing technique that was eventually adopted, and for designing  ...  Separate FIFO memories on the address and data ports of each texture memory bank insure that random page boundary crossings do not significantly degrade the bandwidth available from the dynamic RAMs.  ...  Second, ultimate fill rate is limited by the fragment bandwidth, rather than the primitive bandwidth.  ... 
doi:10.1145/166117.166131 dblp:conf/siggraph/Akeley93 fatcat:kg4wa73ed5hdvbuxfyzaknkzkq
« Previous Showing results 1 — 15 out of 1,122 results