Filters








191,032 Hits in 4.6 sec

Towards estimation error guarantees for distinct values

Moses Charikar, Surajit Chaudhuri, Rajeev Motwani, Vivek Narasayya
2000 Proceedings of the nineteenth ACM SIGMOD-SIGACT-SIGART symposium on Principles of database systems - PODS '00  
For large tables without an index on the column, random sampling appears to be the only scalable approach for estimating the number of distinct values.  ...  We consider the problem of estimating the number of distinct values in a column of a table.  ...  Error Guarantee: There should be an analytic guarantee on the degree of error of the estimator. Low Sampling: The estimator should guarantee small error for relatively low sample sizes.  ... 
doi:10.1145/335168.335230 dblp:conf/pods/CharikarCMN00 fatcat:3im4y3l7nnhlboht7x4f23m2cq

Towards Robust Performance Guarantees for Models Learned from High-Dimensional Data [chapter]

Rui Henriques, Sara C. Madeira
2015 Studies in Big Data  
In these spaces, the performance of models is commonly highly variable and dependent on the target error estimators, data regularities and model properties.  ...  data regularities and learning parameterizations, and iii) guarantee its applicability for different types of models, including classification and descriptive models.  ...  ) or with the poor extensibility of existing approaches towards distinct types of models or flexible data settings.  ... 
doi:10.1007/978-3-319-11056-1_3 fatcat:ayiea5n4sbairdxcywxvsjqdhm

Approximate Query Processing

Surajit Chaudhuri, Bolin Ding, Srikanth Kandula
2017 Proceedings of the 2017 ACM International Conference on Management of Data - SIGMOD '17  
We discuss two promising avenues to pursue towards integrating Approximate Query Processing into data platforms.  ...  Acknowledgements We are deeply thankful to Christian Konig and Vivek Narasayya for reading numerous versions of this document; their feedback has improved this report substantially. REFERENCES  ...  least 30 rows will be sampled for each group (distinct value of i_color).  ... 
doi:10.1145/3035918.3056097 dblp:conf/sigmod/ChaudhuriDK17 fatcat:x2lmtvyhbfdfhpohqlyd23pffq

Summarizing Order Statistics over Data Streams with Duplicates

Ying Zhang, Xuemin Lin, Yidong Yuan, Masaru Kitsuregawa, Xiaofang Zhou, Jeffrey Xu Yu
2007 2007 IEEE 23rd International Conference on Data Engineering  
Relative rank error (or biased) quantile computation techniques have been recently developed in [5, 6, 22] , which aim to give finer rank error guarantees towards heads; that is, enforce the precision  ...  Moreover, to ensure relative rank errors for a give rank r < 1 precise answers are the only possibility; consequently, we always keep the L smallest distinct elements (i.e., L distinct elements with the  ... 
doi:10.1109/icde.2007.369004 dblp:conf/icde/ZhangLYKZY07 fatcat:ntbq3exijbbcfjcqun7tb4km3i

Energy Efficient Schemes for Accuracy-Guaranteed Sensor Data Aggregation Using Scalable Counting

Yao-Chung Fan, Arbee L.P. Chen
2012 IEEE Transactions on Knowledge and Data Engineering  
We focus on having an (", ) accuracy guarantee for computing an aggregate, which ensures that the error in computing the aggregate is within a factor of " with probability (1 À ).  ...  Our schemes using the scalable counting technique efficiently compute the aggregates under a given accuracy guarantee.  ...  The relative error indicates the percentage the estimateŝ will be misestimated on average. The error of the estimate comes from the fact that distinct items share the same bit.  ... 
doi:10.1109/tkde.2011.76 fatcat:il6blgyx7jhr5p7wou4lqjvqsy

Page 1012 of Neural Computation Vol. 6, Issue 5 [page]

1994 Neural Computation  
For the largest models, deviations from the optimal value as little as one-tenth the sampling period will introduce misfit errors greater than o,.  ...  Conversely, if there are more spike models than distinct APs, not only will there be excess computational overhead, but there is no guarantee that each AP will be represented, since some spike functions  ... 

Distinct value estimation on peer-to-peer networks

Zubin Joseph, Gautam Das, Leonidas Fegaras
2008 Proceedings of the 1st ACM international conference on PErvasive Technologies Related to Assistive Environments - PETRA '08  
In this paper, we present a technique to obtain estimations of the number of distinct values matching a query on the network.  ...  However, the sheer scale of these networks has made it difficult to gather statistics that could be used for building new features.  ...  Distinct Value Estimators We discuss two distinct value estimators, the Guaranteed-Error Estimator (GEE) [14] and the Adaptive Estimator [4] .  ... 
doi:10.1145/1389586.1389617 dblp:conf/petra/JosephDF08 fatcat:ybzpdbh425dkrkrvyh2m5qwszi

Duplicate-Insensitive Order Statistics Computation over Data Streams

Ying Zhang, Xuemin Lin, Yidong Yuan, Masaru Kitsuregawa, Xiaofang Zhou, Jeffrey Wu Yu
2010 IEEE Transactions on Knowledge and Data Engineering  
guarantee ǫ in the presence of data duplicates.  ...  Moreover, our techniques may be immediately applied to the heavy hitter problem against distinct elements and to the existing fault-tolerant distributed communication techniques.  ...  error guarantee ǫ for the setting ǫ = 0.02 and (1 − δ) = 0.8.  ... 
doi:10.1109/tkde.2009.68 fatcat:23igwnur4vhs3pukb2qor76uym

Cardinality estimation

Hazar Harmouch, Felix Naumann
2017 Proceedings of the VLDB Endowment  
First, we outline and classify approaches to solve the problem of cardinality estimation -we describe their main idea, error-guarantees, advantages, and disadvantages.  ...  Among the most important types of metadata is the number of distinct values in a column, also known as the zeroth-frequency moment.  ...  However, for the sake of completeness, we briefly evaluated Guaranteed-Error Estimator(GEE) [8] as an example of sampling-based cardinality estimator.  ... 
doi:10.1145/3186728.3164145 fatcat:ia7kzc7yangpbj5sqr3k3zcuvq

Adaptive Channel Estimation For SCFDMA System
English

2014 International Journal of Research and Applications  
Rather, it uses a phase weighting scheme to eliminate the signal fluctuations due to noise and decision errors. The convergence towards the true channel coefficient is guaranteed.  ...  A variable step size based least mean squares (LMS) algorithm is formulated for a single carrier frequency division multiple access (SC-FDMA) system channel estimation (CE).  ...  towards the true channel coefficients, which avoids a fast convergence when the step size is preferred to be small for small output estimation error.  ... 
doi:10.17812/ijra.1.3(22)2014 fatcat:wpw73wshsbguvml7jc43lzpzt4

Robust wide baseline pose estimation from video

Nicola Pellicano, Emanuel Aldea, Sylvie Le Hegarat-Mascle
2016 2016 23rd International Conference on Pattern Recognition (ICPR)  
In this work, we highlight some current limitations of conventional strategies for relative pose estimation in difficult urban scenes.  ...  Robust wide baseline pose estimation is an essential step in the deployment of smart camera networks.  ...  ACKNOWLEDGMENT The authors gratefully acknowledge the support from Regent's Park Mosque for providing access to the site during data collection, and from K. Kiyani.  ... 
doi:10.1109/icpr.2016.7900230 dblp:conf/icpr/PellicanoAH16 fatcat:s5dk4knayrd45olrufvs27a6lq

The extended least-squares and the joint maximum-a-posteriori maximum-likelihood estimation criteria

A. Yeredor, E. Weinstein
1999 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258)  
The Least-Squares (LS) estimation criterion assumes the measured data to be exact, and seeks parameters which minimize the model errors.  ...  We define 'pseudo-linear' models, with which we provide an iterative algorithm for minimization of the XLS criterion.  ...  Thus, in each iteration the value of C,(%, 8 ) is guaranteed not to increase (usually to decrease).  ... 
doi:10.1109/icassp.1999.758273 dblp:conf/icassp/YeredorW99 fatcat:6k3rh43c6fgf3fuhijopbvp2vu

Data streaming algorithms for estimating entropy of network traffic

Ashwin Lall, Vyas Sekar, Mitsunori Ogihara, Jun Xu, Hui Zhang
2006 Proceedings of the joint international conference on Measurement and modeling of computer systems - SIGMETRICS '06/Performance '06  
The first algorithm for entropy estimation is inspired by the structural similarity with the seminal work of Alon et al. for estimating frequency moments, and we provide strong theoretical guarantees on  ...  the error and resource usage.  ...  Muthukrishnan for their useful discussion and for kindly sharing the most recent version of their paper [1] with us. We thank Minho Sung for helping us with the datasets used in this paper.  ... 
doi:10.1145/1140277.1140295 dblp:conf/sigmetrics/LallSOXZ06 fatcat:xywcp4ewtvajzidazf7ixo4e3i

Data streaming algorithms for estimating entropy of network traffic

Ashwin Lall, Vyas Sekar, Mitsunori Ogihara, Jun Xu, Hui Zhang
2006 Performance Evaluation Review  
The first algorithm for entropy estimation is inspired by the structural similarity with the seminal work of Alon et al. for estimating frequency moments, and we provide strong theoretical guarantees on  ...  the error and resource usage.  ...  Muthukrishnan for their useful discussion and for kindly sharing the most recent version of their paper [1] with us. We thank Minho Sung for helping us with the datasets used in this paper.  ... 
doi:10.1145/1140103.1140295 fatcat:ahj2frwztvcsrbod3eqvu25p7q

PrivateClean

Sanjay Krishnan, Jiannan Wang, Michael J. Franklin, Ken Goldberg, Tim Kraska
2016 Proceedings of the 2016 International Conference on Management of Data - SIGMOD '16  
The key insight is to maintain a bipartite graph relating dirty values to clean values and use this graph to estimate biases due to the interaction between cleaning and privacy.  ...  PrivateClean includes a technique for creating private datasets of numerical and discrete-valued attributes, a formalism for privacy-preserving data cleaning, and techniques for answering sum, count, and  ...  Consequently, Direct is more sensitive to errors that "merge" distinct values. terms the estimates from PrivateClean are useful.  ... 
doi:10.1145/2882903.2915248 dblp:conf/sigmod/KrishnanWFGK16 fatcat:rrfg5zrcofd4nphvflu5pmbpxu
« Previous Showing results 1 — 15 out of 191,032 results