For a Gaussian random tensor with independent entries, the expected injective norm is at most sqrt(2r) times the sum of maximal fiber standard deviations plus C r^3 (ln d)^2 times the largest entry magnitude.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.PR 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Injective norm of random tensors with independent entries
For a Gaussian random tensor with independent entries, the expected injective norm is at most sqrt(2r) times the sum of maximal fiber standard deviations plus C r^3 (ln d)^2 times the largest entry magnitude.