1: \begin{abstract}

2:  Understanding the limiting behavior of eigenvalues of random matrices

3:  is the central problem of random matrix theory.  Classical limit

4:  results are known for many models, and there has been significant

5:  recent progress in obtaining more quantitative, non-asymptotic

6:  results.  In this paper, we describe a systematic approach to bounding rates of

7: convergence and proving tail inequalities for the empirical spectral

8: measures of a wide variety of random matrix ensembles.  We illustrate

9: the approach by proving asymptotically almost sure rates of

10: convergence of the empirical spectral measure in the following

11: ensembles:  Wigner matrices,

12: Wishart matrices, Haar-distributed matrices from the compact classical

13: groups, powers of Haar matrices, randomized sums and random

14: compressions of Hermitian matrices, a random matrix model for the

15: Hamiltonians of quantum spin glasses, and finally the complex Ginibre

16: ensemble.  Many of the results appeared previously and are being

17: collected and described here as illustrations of the general method; however, some

18: details (particularly in the Wigner and Wishart cases) are new.

19: 

20: Our approach

21: makes use of techniques from probability in Banach spaces, in

22: particular concentration of measure and bounds for suprema of

23: stochastic processes, in combination with more classical tools from

24: matrix analysis, approximation theory, and Fourier analysis.  It is

25: highly flexible, as evidenced by the broad list of examples.   It is

26: moreover based largely on ``soft'' methods, and

27: involves little hard analysis. 

28: \end{abstract}

29: