AT&T Labs

About: AT&T Labs is a based out in . It is known for research contribution in the topics: Network packet & The Internet. The organization has 1879 authors who have published 5595 publications receiving 483151 citations.

Off-Line Generation of Limited-Use Credit Card Numbers

Abstract: Recently, some credit card companies have introduced limited-use credit card numbers--for example, American Express's single-use card numbers and Visa's gift cards. Such limited-use credit cards limit the exposure of a traditional long-term credit card number, particularly in Internet transactions. These offerings employ an on-line solution, in that a credit card holder must interact with the credit card issuer in order to derive a limited-use token. In this paper, we describe a method for cryptographic off-line generation of limited-use credit card numbers. This has several advantages over the on-line schemes, and it has applications to calling cards as well. We show that there are several trade-offs between security and maintaining the current infrastructure.

Learning dependency translation models as collections of finite-state head transducers

Abstract: The paper defines weighted head transducers, finite-state machines that perform middle-out string transduction. These transducers are strictly more expressive than the special case of standard left-to-right finite-state transducers. Dependency transduction models are then defined as collections of weighted head transducers that are applied hierarchically. A dynamic programming search algorithm is described for finding the optimal transduction of an input string with respect to a dependency transduction model. A method for automatically training a dependency transduction model from a set of input-output example strings is presented. The method first searches for hierarchical alignments of the training examples guided by correlation statistics, and then constructs the transitions of head transducers that are consistent with these alignments. Experimental results are given for applying the training method to translation from English to Spanish and Japanese.

Spanners and emulators with sublinear distance errors

Abstract: Let k ≥ 2 be an integer. We show that any undirected and unweighted graph G = (V, E) on n vertices has a subgraph G' = (V, E') with O(kn1+1/k) edges such that for any two vertices u, v ∈ V, if δG(u, v) = d, then δG'(u, v) = d+O(d1-1/k-1). Furthermore, we show that such subgraphs can be constructed in O(mn1/k) time, where m and n are the number of edges and vertices in the original graph. We also show that it is possible to construct a weighted graph G* = (V, E*) with O(kn1+1/(2k-1)) edges such that for every u, v ∈ V, if δG(u, v) = d, then δ ≤ δG*(u, v) = d + O(d1-1/k-1). These are the first such results with additive error terms of the form o(d), i.e., additive error terms that are sublinear in the distance being approximated.

Probabilistic quorum systems

Abstract: We initiate the study of probabilistic quorum systems, a technique for providing consistency of replicated data with high levels of assurance despite the failure of data servers. We show that this technique offers effective load reduction on servers and high availability. We explore probabilistic quorum systems both for services tolerant of benign server failures and for services tolerant of arbitrary (Byzantine) ones. We also prove bounds on the server load that can be achieved with these techniques. C � 2001 Academic Press