Math @ Duke

Publications [#342383] of Henry Pfister
Papers Published
 Reeves, G; Pfister, HD, The ReplicaSymmetric Prediction for Random Linear Estimation With Gaussian Matrices Is Exact,
Ieee Transactions on Information Theory, vol. 65 no. 4
(April, 2019),
pp. 22522283 [doi]
(last updated on 2019/06/20)
Abstract: © 2019 IEEE. This paper considers the fundamental limit of random linear estimation for i.i.d. signal distributions and i.i.d. Gaussian measurement matrices. Its main contribution is a rigorous characterization of the asymptotic mutual information (MI) and minimum meansquare error (MMSE) in this setting. Under mild technical conditions, our results show that the limiting MI and MMSE are equal to the values predicted by the replica method from statistical physics. This resolves a wellknown problem that has remained open for over a decade.


dept@math.duke.edu
ph: 919.660.2800
fax: 919.660.2821
 
Mathematics Department
Duke University, Box 90320
Durham, NC 277080320

