1998
DOI: 10.1002/(sici)1098-2418(199807)12:4<351::aid-rsa3>3.0.co;2-s
|Get access via publisher |Summarize |Cite
Balancing vectors and Gaussian measures ofn-dimensional convex bodies
Abstract: Let и be the Euclidean norm on R and ␥ the standard Gaussian measure n n Ž . ynr2 y5x5 2 r2
Search citation statements
Paper Sections
Select...
123
47
4
2
Citation Types
3
209
0
4
Year Published
1998
2026
Publication Types
Select...
80
48
18
2
Relationship
0
148
Authors
Journals
Cited by 148 publications
(216 citation statements)
References 0 publications
3
209
0
4
“…This result essentially recovers the best known bound on the Komlós conjecture due to Banaszczyk [4] in an online algorithmic fashion, with the slight defect of requiring a +2-signing option. Furthermore due to the online nature of the algorithm, the algorithm will run in essentially input-sparsity time which is substantially faster than the Gram-Schmidt walk [7] which gives an algorithmic proof of the result of [4] (without the defect of requiring a +2-signing option).…”
Section: Corollary 12 For Any Vectorssupporting
confidence: 80%
“…This result essentially recovers the best known bound on the Komlós conjecture due to Banaszczyk [4] in an online algorithmic fashion, with the slight defect of requiring a +2-signing option. Furthermore due to the online nature of the algorithm, the algorithm will run in essentially input-sparsity time which is substantially faster than the Gram-Schmidt walk [7] which gives an algorithmic proof of the result of [4] (without the defect of requiring a +2-signing option).…”
Section: Corollary 12 For Any Vectorssupporting
confidence: 80%
“…A bound in terms of the maximum degree. If F has maximum degree t, i.e., no point is in more than t sets, then we get disc F = O( √ t log m) by (A) and (B), which recovers the current best bound for this problem, due to Banaszczyk [Ban98]. However, this example is not quite fair, since inequality (2) used in (A) relies on a more general form of Banaszczyk's estimate.…”
Section: Simple Proofs Of Known Discrepancy Boundssupporting
confidence: 61%
“…Suppose that B \ {0} is nonempty, and select an element t from it. By the same argument we had earlier, we have | f (1)…”
Section: The Continuity Argument and A Simple Lower Boundsupporting
confidence: 59%
