Results 131 to 140 of about 232,423 (182)
Some of the next articles are maybe not open access.
Sampling streaming data with replacement
Computational Statistics and Data Analysis, 2007zbMATH Open Web Interface contents unavailable due to conflicting licenses.
Byung-Hoon Park
exaly +4 more sources
Replacement Costs: the Inefficiencies of Sampling with Replacement
Mathematics Magazine, 2005is R* = 16 cm. TABLE 1 gives the value of p for selected values of n. Surprisingly enough, we observe that even when n is equal to 2, which means that the cake is cut into four equal parts, the probability that the knife hits the coin is very high (close to 30%). As expected, p is increasing as n increases and for n = 23, p is just below 1. For n > 24,
E. Murphree
semanticscholar +2 more sources
Determining the number of clusters by sampling with replacement.
Psychological Methods, 2004A split-sample replication criterion originally proposed by J. E. Overall and K. N. Magee (1992) as a stopping rule for hierarchical cluster analysis is applied to multiple data sets generated by sampling with replacement from an original simulated primary data set.
Scott Tonidandel, J. Overall
semanticscholar +3 more sources
XI.—On Fourfold Sampling with and without Replacement
Proceedings of the Royal Society of Edinburgh, 1936When each individual of a population ofNmembers can be classified as being either A orA, and at the same time either B orB, the relative proportions, or probabilities, of the four types AB, AB,AB andABcan be set out in the familiar “fourfold table.”
A. C. Aitken, H. Gonin
semanticscholar +3 more sources
A NOTE ON SAMPLING WITH AND WITHOUT REPLACEMENT
Statistica Neerlandica, 1986AbstractUpper bounds are given for the total variation distance between the distribution statistics of random samples with and without replacement from a finite set.
exaly +2 more sources
An Efficient Method for Weighted Sampling without Replacement
SIAM Journal on Computing, 1980In this note, an efficient method for weighted sampling of K objects without replacement from a population of n objects is proposed. The method requires $O(K\log n)$ additions and comparisons, and $O(K)$ multiplications and random number generations while the method proposed by Fagin and Price requires $O(Kn)$ additions and comparisons, and $O(K ...
C. K. Wong, Malcolm C. Easton
exaly +2 more sources
Recursive Speculative Decoding: Accelerating LLM Inference via Sampling Without Replacement
arXiv.orgSpeculative decoding is an inference-acceleration method for large language models (LLMs) where a small language model generates a draft-token sequence which is further verified by the target LLM in parallel.
Wonseok Jeon +5 more
semanticscholar +1 more source
Sampling with Partial Replacement and Stratification
Forestry sciences, 1994Sampling with Partial Replacement (SPR) is extended to double sampling for stratification on three occasions based on the two-occasion case presented by Bickford et al. (1963).
C. Scott, M. Köhl
semanticscholar +1 more source
Accelerated Sampling from Masked Diffusion Models via Entropy Bounded Unmasking
arXiv.orgRecent masked diffusion models (MDMs) have shown competitive performance compared to autoregressive models (ARMs) for language modeling. While most literature has focused on performance enhancing sampling procedures, efficient sampling from MDMs has been
Heli Ben-Hamu +4 more
semanticscholar +1 more source