J
Jiashun Jin
Researcher at Carnegie Mellon University
Publications - 89
Citations - 5341
Jiashun Jin is an academic researcher from Carnegie Mellon University. The author has contributed to research in topics: Feature selection & Computer science. The author has an hindex of 37, co-authored 81 publications receiving 4706 citations. Previous affiliations of Jiashun Jin include Purdue University & Stanford University.
Papers
More filters
Journal ArticleDOI
Higher criticism for detecting sparse heterogeneous mixtures
David L. Donoho,Jiashun Jin +1 more
TL;DR: In this paper, higher criticism is used to test whether n normal means are all zero versus the alternative that a small fraction of nonzero means is nonzero, and it is shown that higher criticism works well over a range of non-Gaussian cases.
Journal ArticleDOI
Fast community detection by score
TL;DR: A theoretic framework is developed where it is shown that under mild conditions, the SCORE stably yields successful community detection and is much more satisfactory than those by the classical spectral methods.
Journal ArticleDOI
The Non-Gaussian Cold Spot in the 3 Year Wilkinson Microwave Anisotropy Probe Data
TL;DR: In this paper, a non-Gaussian cold spot was detected in wavelet space in the WMAP 1-year data, and was detected again in the coadded WMAP 3 -year data at the same position (b = 57 ◦,l = 209 ◦ ) and size in the sky (� 10 ◦ ).
Journal ArticleDOI
Higher criticism thresholding: Optimal feature selection when useful features are rare and weak.
David L. Donoho,Jiashun Jin +1 more
TL;DR: In the most challenging RW settings, HCT uses an unconventionally low threshold, which keeps the missed-feature detection rate under better control than FDRT and yields a classifier with improved misclassification performance.
Journal ArticleDOI
Innovated Higher Criticism for Detecting Sparse Signals in Correlated Noise
Peter A. Hall,Jiashun Jin +1 more
TL;DR: In this paper, it was shown that correlation can be used to improve the performance of higher-criticism in the presence of correlated signals. But, it was also shown that the case of independent noise is the most difficult of all, from a statistical viewpoint, and that more accurate signal detection can be obtained when correlation is present.