Near-Optimality of Linear Recovery from Indirect Observations

We consider the problem of recovering linear image of a signal known to belong to a given convex compact set from indirect observation of corrupted by random noise with finite covariance matrix. It is shown that under some assumptions on (satisfied, e.g., when is the intersection of concentric ellipsoids/elliptic cylinders, or the unit ball of the spectral norm in the space of matrices) and on the norm used to measure the recovery error (satisfied, e.g., by -norms, , on and by the nuclear norm on the space of matrices), one can build, in a computationally efficient manner, a "presumably good" linear in observations estimate, and that in the case of zero mean Gaussian observation noise, this estimate is near-optimal among all (linear and nonlinear) estimates in terms of its worst-case, over , expected -loss. These results form an essential extension of those in our paper arXiv:1602.01355, where the assumptions on were more restrictive, and the norm was assumed to be the Euclidean one. In addition, we develop near-optimal estimates for the case of "uncertain-but-bounded" noise, where all we know about is that it is bounded in a given norm by a given . Same as in arXiv:1602.01355, our results impose no restrictions on and . This arXiv paper slightly strengthens the journal publication Juditsky, A., Nemirovski, A. "Near-Optimality of Linear Recovery from Indirect Observations," Mathematical Statistics and Learning 1:2 (2018), 171-225.
View on arXiv