Averaging m independent SBM samples before spectral clustering improves the recovery condition from (a-b)^2/(a+b) >= C log(2/gamma) to the same divided by m.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
SBM With Multiple Samples: Improved Spectral Recovery
Averaging m independent SBM samples before spectral clustering improves the recovery condition from (a-b)^2/(a+b) >= C log(2/gamma) to the same divided by m.