Almost Everywhere Matrix Recovery
نویسندگان
چکیده
Matrix recovery is raised in many areas. In this paper, we build up a framework for almost everywhere matrix recovery which means to recover almost all the P ∈ M ⊂ F from Tr(AjP ), j = 1, . . . , N where Aj ∈ Vj ⊂ F. We mainly focus on the following question: how many measurements are needed to recover almost all the matrices in M? For the case where both M and Vj are algebraic varieties, we use the tools from algebraic geometry to study the question and present some results to address it under many different settings.
منابع مشابه
Active Learning of Multi-Index Function Models
We consider the problem of actively learning multi-index functions of the form f(x) = g(Ax) = ∑k i=1 gi(a T i x) from point evaluations of f . We assume that the function f is defined on an `2-ball in R, g is twice continuously differentiable almost everywhere, and A ∈ Rk×d is a rank k matrix, where k d. We propose a randomized, active sampling scheme for estimating such functions with uniform ...
متن کاملApproximate Subgradients and Coderivatives in Rn
We show that in two dimensions or higher, the Mordukhovich-loffe approximate subdifferential and Clarke subdifferential may differ almost everywhere for real-valued Lipschitz functions. Uncountably many Frechet differentiable vector-valued Lipschitz functions differing by more than constants can share the same Mordukhovich-Ioffe coderivatives. Moreover, the approximate Jacobian associated with ...
متن کاملAlmost Everywhere Positivity of the Lyapunov Exponent for the Doubling Map
We show that discrete one-dimensional Schrödinger operators on the half-line with ergodic potentials generated by the doubling map on the circle, Vθ(n) = f(2 θ), may be realized as the half-line restrictions of a nondeterministic family of whole-line operators. As a consequence, the Lyapunov exponent is almost everywhere positive and the absolutely continuous spectrum is almost surely empty. Co...
متن کاملOn absolute matrix summability of orthogonal series
In this paper, we prove two theorems on ∣A∣k, 1 ≤ k ≤ 2, summability of orthogonal series. The first one gives a sufficient condition under which an orthogonal series is absolutely summable almost everywhere, and the second one, is a general theorem, which also gives a sufficient condition so that an orthogonal series is absolutely summable almost everywhere, but it involves a positive numerica...
متن کاملDivergence of Random Power Series
then almost all series (1) diverge almost everywhere on I zl = 1 . Here almost all refers to the set of t (in the usual Lebesgue sense), while almost everywhere refers to the set of z on the circumference of the unit circle (again in the usual sense) . Only recently [1] was it observed that some nontrivial interesting assertions similar to the above with almost everywhere replaced by everywhere...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
- CoRR
دوره abs/1707.09112 شماره
صفحات -
تاریخ انتشار 2017