Four Deviations Suffice for Rank 1 Matrices

We prove a matrix discrepancy bound that strengthens the famous Kadison-Singer result of Marcus, Spielman, and Srivastava. Consider any independent scalar random variables $\xi_1, \ldots, \xi_n$ with finite support, e.g. $\{ \pm 1 \}$ or $\{ 0,1 \}$-valued random variables, or some combination thereof. Let $u_1, \dots, u_n \in \mathbb{C}^m$ and $$ \sigma^2 = \left\| \sum_{i=1}^n \text{Var}[ \xi_i ] (u_i u_i^{*})^2 \right\|. $$ Then there exists a choice of outcomes $\varepsilon_1,\ldots,\varepsilon_n$ in the support of $\xi_1, \ldots, \xi_n$ s.t. $$ \left \|\sum_{i=1}^n \mathbb{E} [ \xi_i] u_i u_i^* - \sum_{i=1}^n \varepsilon_i u_i u_i^* \right \| \leq 4 \sigma. $$ A simple consequence of our result is an improvement of a Lyapunov-type theorem of Akemann and Weaver.

[1]  Rudolf Ahlswede,et al.  Strong converse for identification via quantum channels , 2000, IEEE Trans. Inf. Theory.

[2]  He Sun.  Constructing Linear-Sized Spectral Sparsification in Almost-Linear Time , 2017 .

[3]  Shachar Lovett,et al.  The Gram-Schmidt walk: a cure for the Banaszczyk blues , 2017, STOC.

[4]  Nikhil Bansal,et al.  An Algorithm for Komlós Conjecture Matching Banaszczyk's Bound , 2019, SIAM J. Comput..

[5]  Nicholas Weaver.  The Kadison-Singer problem in discrepancy theory , 2004, Discret. Math..

[6]  B Cohen Michael.  Ramanujan Graphs in Polynomial Time , 2016 .

[7]  M. Rudelson.  Random Vectors in the Isotropic Position , 1996, math/9608208.

[8]  Paul D. Seymour,et al.  The roots of the independence polynomial of a clawfree graph , 2007, J. Comb. Theory B.

[9]  J. Borcea,et al.  Applications of stable polynomials to mixed determinants: Johnson's conjectures, unimodality, and symmetrized Fischer products , 2006, math/0607755.

[10]  R. J. Gregorac,et al.  CorrigendumObreschkoff's theorem revisited: What convex sets are contained in the set of hyperbolic polynomials?: Journal of pure and applied algebra 81 (3) (1992) 269–278 , 1994 .

[11]  Shang-Hua Teng,et al.  Nearly-linear time algorithms for graph partitioning, graph sparsification, and solving linear systems , 2003, STOC '04.

[12]  Nikhil Srivastava,et al.  Graph Sparsification by Effective Resistances , 2011, SIAM J. Comput..

[13]  Thomas Rothvoß,et al.  Approximating Bin Packing within O(log OPT * Log Log OPT) Bins , 2013, 2013 IEEE 54th Annual Symposium on Foundations of Computer Science.

[14]  Aleksandar Nikolov,et al.  Balancing Vectors in Any Norm , 2018, 2018 IEEE 59th Annual Symposium on Foundations of Computer Science (FOCS).

[15]  Nima Anari,et al.  The Kadison-Singer Problem for Strongly Rayleigh Measures and Applications to Asymmetric TSP , 2014, ArXiv.

[16]  Joel A. Tropp,et al.  User-Friendly Tail Bounds for Sums of Random Matrices , 2010, Found. Comput. Math..

[17]  H. Fell.  On the zeros of convex combinations of polynomials. , 1980 .

[18]  J. Spencer.  Six standard deviations suffice , 1985 .

[19]  Nikhil Srivastava,et al.  Interlacing Families IV: Bipartite Ramanujan Graphs of All Sizes , 2015, 2015 IEEE 56th Annual Symposium on Foundations of Computer Science.

[20]  D. Spielman,et al.  Interlacing Families II: Mixed Characteristic Polynomials and the Kadison-Singer Problem , 2013, 1306.3969.

[21]  Nik Weaver,et al.  A Lyapunov‐type theorem from Kadison–Singer , 2013, 1308.5276.

[22]  M. Murty.  Ramanujan Graphs , 1965 .

[23]  Noga Alon,et al.  The Probabilistic Method , 2015, Fundamentals of Ramsey Theory.

[24]  Amin Saberi,et al.  Approximating the Largest Root and Applications to Interlacing Families , 2017, SODA.

[25]  Nikhil Bansal,et al.  Constructive Algorithms for Discrepancy Minimization , 2010, 2010 IEEE 51st Annual Symposium on Foundations of Computer Science.

[26]  Yin Tat Lee,et al.  An SDP-based algorithm for linear-sized spectral sparsification , 2017, STOC.

[27]  Nikhil Srivastava,et al.  Interlacing Families I: Bipartite Ramanujan Graphs of All Degrees , 2013, 2013 IEEE 54th Annual Symposium on Foundations of Computer Science.

[28]  Thomas Rothvoß,et al.  Constructive Discrepancy Minimization for Convex Sets , 2014, 2014 IEEE 55th Annual Symposium on Foundations of Computer Science.

[29]  Bernard Chazelle,et al.  The Discrepancy Method , 1998, ISAAC.

[30]  C. Akemann,et al.  Lyapunov theorems for operator algebras , 1991 .

[31]  Shachar Lovett,et al.  Constructive Discrepancy Minimization by Walking on the Edges , 2012, FOCS.

[32]  D. Wagner,et al.  Multivariate stable polynomials: theory and applications , 2009, 0911.3569.

[33]  Nikhil Bansal,et al.  Algorithmic discrepancy beyond partial coloring , 2016, STOC.

[34]  Ding‐Zhu Du,et al.  Wiley Series in Discrete Mathematics and Optimization , 2014 .

[35]  Nikhil Srivastava,et al.  Twice-Ramanujan Sparsifiers , 2012, SIAM J. Comput..