Showing papers by "AT&T Labs published in 2000"

Nonlinear dimensionality reduction by locally linear embedding.

Abstract: Many areas of science depend on exploratory data analysis and visualization. The need to analyze large amounts of multivariate data raises the fundamental problem of dimensionality reduction: how to discover compact representations of high-dimensional data. Here, we introduce locally linear embedding (LLE), an unsupervised learning algorithm that computes low-dimensional, neighborhood-preserving embeddings of high-dimensional inputs. Unlike clustering methods for local dimensionality reduction, LLE maps its inputs into a single global coordinate system of lower dimensionality, and its optimizations do not involve local minima. By exploiting the local symmetries of linear reconstructions, LLE is able to learn the global structure of nonlinear manifolds, such as those generated by images of faces or documents of text.

Ultra-wide bandwidth time-hopping spread-spectrum impulse radio for wireless multiple-access communications

Abstract: Attractive features of time-hopping spread-spectrum multiple-access systems employing impulse signal technology are outlined, and emerging design issues are described. Performance of such communications systems in terms of achievable transmission rate and multiple-access capability are estimated for both analog and digital data modulation formats under ideal multiple-access channel conditions.

Simple proof of security of the BB84 quantum key distribution protocol

Abstract: We prove that the 1984 protocol of Bennett and Brassard (BB84) for quantum key distribution is secure. We first give a key distribution protocol based on entanglement purification, which can be proven secure using methods from Lo and Chau's proof of security for a similar protocol. We then show that the security of this protocol implies the security of BB84. The entanglement purification based protocol uses Calderbank-Shor-Steane codes, and properties of these codes are used to remove the use of quantum computation from the Lo-Chau protocol.

Fair end-to-end window-based congestion control

Abstract: In this paper, we demonstrate the existence of fair end-to-end window-based congestion control protocols for packet-switched networks with first come-first served routers. Our definition of fairness generalizes proportional fairness and includes arbitrarily close approximations of max-min fairness. The protocols use only information that is available to end hosts and are designed to converge reasonably fast. Our study is based on a multiclass fluid model of the network. The convergence of the protocols is proved using a Lyapunov function. The technical challenge is in the practical implementation of the protocols.

BoosTexter: A Boosting-based Systemfor Text Categorization

Abstract: This work focuses on algorithms which learn from examples to perform multiclass text and speech categorization tasks. Our approach is based on a new and improved family of boosting algorithms. We describe in detail an implementation, called BoosTexter, of the new boosting algorithms for text categorization tasks. We present results comparing the performance of BoosTexter and a number of other text-categorization algorithms on a variety of tasks. We conclude by describing the application of our system to automatic call-type identification from unconstrained spoken customer responses.

An open graph visualization system and its applications to software engineering

Abstract: SUMMARY We describe a package of practical tools and libraries for manipulating graphs and their drawings. Our design, which aimed at facilitating the combination of the package components with other tools, includes stream and event interfaces for graph operations, high-quality static and dynamic layout algorithms, and the ability to handle sizable graphs. We conclude with a description of the applications of this package to a variety of software engineering tools. Copyright c 1999 John Wiley & Sons, Ltd.

Interaction and outeraction: instant messaging in action

Abstract: We discuss findings from an ethnographic study of instant messaging (IM) in the workplace and its implications for media theory. We describe how instant messaging supports a variety of informal communication tasks. We document the affordances of IM that support flexible, expressive communication. We describe some unexpected uses of IM that highlight aspects of communication which are not part of current media theorizing. They pertain to communicative processes people use to connect with each other and to manage communication, rather than to information exchange. We call these processes "outeraction". We discuss how outeractional aspects of communication affect media choice and patterns of media use.

Internet traffic engineering by optimizing OSPF weights

Abstract: Open shortest path first (OSPF) is the most commonly used intra-domain Internet routing protocol. Traffic flow is routed along shortest paths, splitting flow at nodes where several outgoing links are on shortest paths to the destination. The weights of the links, and thereby the shortest path routes, can be changed by the network operator. The weights could be set proportional to their physical distances, but often the main goal is to avoid congestion, i.e., overloading of links, and the standard heuristic recommended by Cisco is to make the weight of a link inversely proportional to its capacity. Our starting point was a proposed AT&T WorldNet backbone with demands projected from previous measurements. The desire was to optimize the weight setting based on the projected demands. We showed that optimizing the weight settings for a given set of demands is NP-hard, so we resorted to a local search heuristic. Surprisingly it turned out that for the proposed AT&T WorldNet backbone, we found weight settings that performed within a few percent from that of the optimal general routing where the flow for each demand is optimally distributed over all paths between source and destination. This contrasts the common belief that OSPF routing leads to congestion and it shows that for the network and demand matrix studied we cannot get a substantially better load balancing by switching to the proposed more flexible multi-protocol label switching (MPLS) technologies. Our techniques were also tested on synthetic internetworks, based on a model of Zegura et al., (1996), for which we did not always get quite as close to the optimal general routing.

Multiagent Systems: A Survey from a Machine Learning Perspective

Abstract: Distributed Artificial Intelligence (DAI) has existed as a subfield of AI for less than two decades. DAI is concerned with systems that consist of multiple independent entities that interact in a domain. Traditionally, DAI has been divided into two sub-disciplines: Distributed Problem Solving (DPS) focuses on the information management aspects of systems with several components working together towards a common goals Multiagent Systems (MAS) deals with behavior management in collections of several independent entities, or agents. This survey of MAS is intended to serve as an introduction to the field and as an organizational framework. A series of general multiagent scenarios are presented. For each scenario, the issues that arise are described along with a sampling of the techniques that exist to deal with them. The presented techniques are not exhaustive, but they highlight how multiagent systems can be and have been used to build complex systems. When options exist, the techniques presented are biased towards machine learning approaches. Additional opportunities for applying machine learning to MAS are highlighted and robotic soccer is presented as an appropriate test bed for MAS. This survey does not focus exclusively on robotic systems. However, we believe that much of the prior research in non-robotic MAS is relevant to robotic MAS, and we explicitly discuss several robotic MAS, including all of those presented in this issue.

A random graph model for massive graphs

Abstract: We propose a random graph model which is a special case of sparse random graphs with given degree sequences. This model involves only a small number of parameters, called logsize and log-log growth rate. These parameters capture some universal characteristics of massive graphs. Furthermore, from these parameters, various properties of the graph can be derived. For example, for certain ranges of the parameters, we will compute the expected distribution of the sizes of the connected components which almost surely occur with high probability. We will illustrate the consistency of our model with the behavior of some massive graphs derived from data in telecommunications. We will also discuss the threshold function, the giant component, and the evolution of random graphs in this model.

A differential detection scheme for transmit diversity

Abstract: We present a transmission scheme for exploiting diversity given by two transmit antennas when neither the transmitter nor the receiver has access to channel state information. The new detection scheme can use equal energy constellations and encoding is simple. At the receiver, decoding is achieved with low decoding complexity. The transmission provides full spatial diversity and requires no channel state side information at the receiver. The scheme can be considered as the extension of differential detection schemes to two transmit antennas.

Logistic Regression, AdaBoost and Bregman Distances

Abstract: We give a unified account of boosting and logistic regression in which each learning problem is cast in terms of optimization of Bregman distances. The striking similarity of the two problems in this framework allows us to design and analyze algorithms for both simultaneously, and to easily adapt algorithms designed for one problem to the other. For both problems, we give new algorithms and explain their potential advantages over existing methods. These algorithms are iterative and can be divided into two types based on whether the parameters are updated sequentially (one at a time) or in parallel (all at once). We also describe a parameterized family of algorithms that includes both a sequential- and a parallel-update algorithm as special cases, thus showing how the sequential and parallel approaches can themselves be unified. For all of the algorithms, we give convergence proofs using a general formalization of the auxiliary-function proof technique. As one of our sequential-update algorithms is equivalent to AdaBoost, this provides the first general proof of convergence for AdaBoost. We show that all of our algorithms generalize easily to the multiclass case, and we contrast the new algorithms with the iterative scaling algorithm. We conclude with a few experimental results with synthetic data that highlight the behavior of the old and newly proposed algorithms in different settings.

Pilot-symbol-aided channel estimation for OFDM in wireless systems

Abstract: In this paper, we investigate pilot-symbol-aided parameter estimation for orthogonal frequency division multiplexing (OFDM) systems. We first derive a minimum mean-square error (MMSE) pilot-symbol-aided parameter estimator. Then, we discuss a robust implementation of the pilot-symbol-aided estimator that is insensitive to channel statistics. From the simulation results, the required signal-to-noise ratios (SNRs) for a 10% word error rate (WER) are 6.8 dB and 7.3 dB for the typical urban (TU) channels with 40 Hz and 200 Hz Doppler frequencies, respectively, and they are 8 dB and 8.3 dB for the hilly-terrain (HT) channels with 40 Hz and 200 Hz Doppler frequencies, respectively. Compared with the decision-directed parameter estimator, the pilot-symbol-aided estimator is highly robust to Doppler frequency for dispersive fading channels with noise impairment even though it has some performance degradation for systems with lower Doppler frequencies.

Convergence Results for Single-Step On-PolicyReinforcement-Learning Algorithms

Abstract: An important application of reinforcement learning (RL) is to finite-state control problems and one of the most difficult problems in learning for control is balancing the exploration/exploitation tradeoff. Existing theoretical results for RL give very little guidance on reasonable ways to perform exploration. In this paper, we examine the convergence of single-step on-policy RL algorithms for control. On-policy algorithms cannot separate exploration from learning and therefore must confront the exploration problem directly. We prove convergence results for several related on-policy algorithms with both decaying exploration and persistent exploration. We also provide examples of exploration strategies that can be followed during learning that result in convergence to both optimal values and optimal policies.

Bounds on Error Expectation for Support Vector Machines

Abstract: We introduce the concept of span of support vectors (SV) and show that the generalization ability of support vector machines (SVM) depends on this new geometrical concept. We prove that the value of the span is always smaller (and can be much smaller) than the diameter of the smallest sphere containing the support vectors, used in previous bounds (Vapnik, 1998). We also demonstate experimentally that the prediction of the test error given by the span is very accurate and has direct application in model selection (choice of the optimal parameters of the SVM).

Implementing a distributed firewall

Abstract: Conventional rewalls rely on topology restrictions and controlled network entry points to enforce traAEc ltering. Furthermore, a rewall cannot lter traAEc it does not see, so, e ectively, everyone on the protected side is trusted. While this model has worked well for small to medium size networks, networking trends such as increased connectivity, higher line speeds, extranets, and telecommuting threaten to make it obsolete. To address the shortcomings of traditional rewalls, the concept of a \distributed rewall" has been proposed. In this scheme, security policy is still centrally de ned, but enforcement is left up to the individual endpoints. IPsec may be used to distribute credentials that express parts of the overall network policy. Alternately, these credentials may be obtained through out-of-band means. In this paper, we present the design and implementation of a distributed rewall using the KeyNote trust management system to specify, distribute, and resolve policy, and OpenBSD, an open source UNIX operating system.

XMill: an efficient compressor for XML data

Abstract: We describe a tool for compressing XML data, with applications in data exchange and archiving, which usually achieves about twice the compression ratio of gzip at roughly the same speed. The compressor, called XMill, incorporates and combines existing compressors in order to apply them to heterogeneous XML data: it uses zlib, the library function for gzip, a collection of datatype specific compressors for simple data types, and, possibly, user defined compressors for application specific data types.

Rapid ethnography: time deepening strategies for HCI field research

Abstract: Field research methods are useful in the many aspects of Human-Computer Interaction research, including gathering user requirements, understanding and developing user models, and new product evaluation and iterative design. Due to increasingly short product realization cycles, there has been growing interestth in more time efficient methods, including rapid prototyping methods and various usability inspection techniques. This paper will introduce "rapid ethnography," which is a collection of field methods intended to provide a reasonable understanding of users and their activities given significant time pressures and limited time in the field.. The core elements include limiting or constraining the research focus and scope, using key informants, capturing rich field data by using multiple observers and interactive observation techniques, and collaborative qualitative data analysis. A short case study illustrating the important characteristics of rapid ethnography will also be presented.

Characteristics of scalability and their impact on performance

Abstract: is a desirable attribute of a network, system, or process. Poor scalability can result in poor system performance, necessitating the reengineering or duplication of systems. While scalability is valued, its characteristics and the characteristics that undermine it are usually only apparent from the context. Here, we attempt to define different aspects of scalability, such as structural scalability and load scalability. Structural scalability is the ability of a system to expand in a chosen dimension without major modifications to its architecture. Load scalability is the ability of a system to perform gracefully as the offered traffic increases. It is argued that systems with poor load scalability may exhibit it because they repeatedly engage in wasteful activity, because they are encumbered with poor scheduling algorithms, because they cannot fully take advantage of parallelism, or because they are algorithmically inefficient. We qualitatively illustrate these concepts with classical examples from the literature of operating systems and local area networks, as well as an example of our own. Some of these are accompanied by rudimentary delay analysis.

Influence sets based on reverse nearest neighbor queries

Abstract: Inherent in the operation of many decision support and continuous referral systems is the notion of the “influence” of a data point on the database. This notion arises in examples such as finding the set of customers affected by the opening of a new store outlet location, notifying the subset of subscribers to a digital library who will find a newly added document most relevant, etc. Standard approaches to determining the influence set of a data point involve range searching and nearest neighbor queries.In this paper, we formalize a novel notion of influence based on reverse neighbor queries and its variants. Since the nearest neighbor relation is not symmetric, the set of points that are closest to a query point (i.e., the nearest neighbors) differs from the set of points that have the query point as their nearest neighbor (called the reverse nearest neighbors). Influence sets based on reverse nearest neighbor (RNN) queries seem to capture the intuitive notion of influence from our motivating examples.We present a general approach for solving RNN queries and an efficient R-tree based method for large data sets, based on this approach. Although the RNN query appears to be natural, it has not been studied previously. RNN queries are of independent interest, and as such should be part of the suite of available queries for processing spatial and multimedia data. In our experiments with real geographical data, the proposed method appears to scale logarithmically, whereas straightforward sequential scan scales linearly. Our experimental study also shows that approaches based on range searching or nearest neighbors are ineffective at finding influence sets of our interest.

Increasing data rate over wireless channels

Abstract: Space-time coding (STC) is a new coding/signal processing framework for wireless communication systems with multiple transmit and multiple receive antennas. This new framework has the potential of dramatically improve the capacity and data rates. In addition, this framework presents the best trade-off between spectral efficiency and power consumption. ST codes (designed so far) come in two different types. ST trellis codes offer the maximum possible diversity gain and a coding gain without any sacrifice in the transmission bandwidth. The decoding of these codes, however, would require the use of a vector form of the Viterbi decoder. Space-time block codes (STBCs) offer a much simpler may of obtaining transmit diversity without any sacrifice in bandwidth and without requiring huge decoding complexity. In fact, the structure of the STBCs is such that it allows for very simple signal processing (linear combining) for encoding/decoding, differential encoding/detection, and interference cancellation. This new signal processing framework offered by ST codes can be used to enhance the data rate and/or capacity in various wireless applications. That is the reason many of these STC ideas have already found their way to some of the current third-generation wireless systems standards.

A quasi-orthogonal space-time block code

Abstract: It has been shown that a complex orthogonal design which provides full diversity and full transmission rate for a space-time block code is not possible for more than two antennas. Previous attempts have been concentrated in generalizing orthogonal designs which provide space-time block codes with full diversity and a high transmission rate. In this work, we design rate one codes which are quasi-orthogonal and provide partial diversity. The decoder of the proposed codes works with pairs of transmitted symbols instead of single symbols.

Stable Internet routing without global coordination

Abstract: The Border Gateway Protocol (BGP) allows an autonomous system (AS) to apply diverse local policies for selecting routes and propagating reachability information to other domains. However, BGP permits ASes to have conflicting policies that can lead to routing instability. This paper proposes a set of guidelines for an AS to follow in setting its routing policies, without requiring coordination with other ASes. Our approach exploits the Internet's hierarchical structure and the commercial relationships between ASes to impose a partial order on the set of routes to each destination. The guidelines conform to conventional traffic-engineering practices of ISPs, and provide each AS with significant flexibility in selecting its local policies. Furthermore, the guidelines ensure route convergence even under changes in the topology and routing policies. Drawing on a formal model of BGP, we prove that following our proposed policy guidelines guarantees route convergence. We also describe how our methodology can be applied to new types of relationships between ASes, how to verify the hierarchical AS relationships, and how to realize our policy guidelines. Our approach has significant practical value since it preserves the ability of each AS to apply complex local policies without divulging its BGP configurations to others.

On network-aware clustering of Web clients

Abstract: Being able to identify the groups of clients that are responsible for a significant portion of a Web site's requests can be helpful to both the Web site and the clients. In a Web application, it is beneficial to move content closer to groups of clients that are responsible for large subsets of requests to an origin server. We introduce clusters---a grouping of clients that are close together topologically and likely to be under common administrative control. We identify clusters using a ``network-aware" method, based on information available from BGP routing table snapshots.

Beyond 3G: wideband wireless data access based on OFDM and dynamic packet assignment

Abstract: The rapid growth of wireless voice subscribers, the growth of the Internet, and the increasing use of portable computing devices suggest that wireless Internet access will rise rapidly over the next few years. Rapid progress in digital and RF technology is making possible highly compact and integrated terminal devices, and the introduction of sophisticated wireless data software is making wireless Internet access more user-friendly and providing more value. Transmission rates are currently only about 10 kb/s for large cell systems. Third-generation wireless access such as WCDMA and the evolution of second-generation systems such as TDMA IS-136+, EDGE, and CDMA IS-95 will provide nominal bit rates of 50-384 kb/s in macrocellular systems. This article discusses packet data transmission rates of 2-5 Mb/s in macrocellular environments and up to 10 Mb/s in microcellular and indoor environments as a complementary service to evolving second- and third-generation wireless systems. Dynamic packet assignment for high-efficiency resource management and packet admission; OFDM at the physical layer with interference suppression, space-time coding, and frequency diversity; as well as smart antennas to obtain good power and spectral efficiency are discussed in this proposal. Flexible allocation of both large and small resources also permits provisioning of services for different delay and throughput requirements.

SilkRoute: trading between relations and XML

Abstract: XML is the standard format for data exchange between inter-enterprise applications on the Internet. To facilitate data exchange, industry groups define public document type definitions (DTDs) that specify the format of the XML data to be exchanged between their applications. In this paper, we address the problem of automating the conversion of relational data into XML. We describe SilkRoute, a general , dynamic , and efficient tool for viewing and querying relational data in XML. SilkRoute is general, because it can express mappings of relational data into XML that conforms to arbitrary DTDs. We call these mappings views . Applications express the data they need as an XML-QL query over the view. SilkRoute is dynamic, because it only materializes the fragment of an XML view needed by an application, and it is efficient, because it fully exploits the underlying RDBMs query engine whenever data items in an XML view need to be materialized.

Inference for Deterministic Simulation Models: The Bayesian Melding Approach

Abstract: Deterministic simulation models are used in many areas of science, engineering, and policy making. Typically, these are complex models that attempt to capture underlying mechanisms in considerable detail, and they have many user-specified inputs. The inputs are often specified by some form of trial-and-error approach in which plausible values are postulated, the corresponding outputs inspected, and the inputs modified until plausible outputs are obtained. Here we address the issue of more formal inference for such models. A probabilistic approach, called Bayesian synthesis, was shown to suffer from the Borel paradox, according to which the results can depend on the parameterization of the model. We propose a modified approach, called Bayesian melding which takes into full account information and uncertainty about both inputs and outputs to the model, while avoiding the Borel paradox. This is done by recognizing the existence of two priors, one implicit and one explicit, on each input and output; ...

Publius: a robust, tamper-evident, censorship-resistant web publishing system

Abstract: We describe a system that we have designed and implemented for publishing content on the web. Our publishing scheme has the property that it is very difficult for any adversary to censor or modify the content. In addition, the identity of the publisher is protected once the content is posted. Our system differs from others in that we provide tools for updating or deleting the published content, and users can browse the content in the normal point and click manner using a standard web browser and a client-side proxy that we provide. All of our code is freely available.

Experiments with random projection

Abstract: Recent theoretical work has identified random projection as a promising dimensionality reduction technique for learning mixtures of Gaussians. Here we summarize these results and illustrate them by a wide variety of experiments on synthetic and real data.