Inference Under Information Constraints I: Lower Bounds From Chi-Square Contraction
نویسندگان
چکیده
منابع مشابه
Chi-square lower bounds
The information inequality has been shown to be an effective tool for providing lower bounds for the minimax risk. Bounds based on the chisquare distance can sometimes offer a considerable improvement especially when applied iteratively. This paper compares these two methods in four examples including the bounded normal mean problem as well as obervations from a Poisson distribution.
متن کاملContraction and Treewidth Lower Bounds
Edge contraction is shown to be a useful mechanism to improve lower bound heuristics for treewidth. A successful lower bound for treewidth is the degeneracy: the maximum over all subgraphs of the minimum degree. The degeneracy is polynomial time computable. We introduce the notion of contraction degeneracy: the maximum over all minors of the minimum degree. We show that the contraction degenera...
متن کاملGeometric Lower Bounds for Distributed Parameter Estimation under Communication Constraints
We consider parameter estimation in distributed networks, where each node in the network observes an independent sample from an underlying distribution and has k bits to communicate its sample to a centralized processor which computes an estimate of a desired parameter of the distribution. We develop lower bounds for the minimax risk of estimating the underlying parameter under squared l2 loss ...
متن کاملBounds for Stop-Loss Premium Under Restrictions on the Chi-Square Statistic
Stop-loss reinsurance is one type of reinsurance contract that has attracted recent attention. In the simplest form of this contract, a reinsurer agrees to pay all losses of the insurer in excess of an agreed limit. This paper concerns the computation of bounds on the stop{loss premium when the loss distribution is unknown, but information about past claim experience is available in the form of...
متن کاملCommunication Lower Bounds for Tensor Contraction Algorithms
Contractions of nonsymmetric tensors are reducible to matrix multiplication, however, ‘fully symmetric contractions’ in which the tensors are symmetric and the result is symmetrized can be done with fewer operations. The ‘direct evaluation algorithm’ for fully symmetric contractions exploits equivalence between terms in the contraction equation to obtain a lower computation cost than the cost a...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: IEEE Transactions on Information Theory
سال: 2020
ISSN: 0018-9448,1557-9654
DOI: 10.1109/tit.2020.3028440