Papers.

Curriculum development related to information security policies and procedures

Learning structured prediction models: a large margin approach

Near-optimal sensor placements in Gaussian processes

When monitoring spatial phenomena, which are often modeled as Gaussian Processes (GPs), choosing sensor locations is a fundamental task. A common strategy is to place sensors at the points of highest entropy (variance) in the GP model. We propose a mutual information criteria, and show that it produces better placements. Furthermore, we prove that finding the configuration that maximizes mutual information is NP-complete. To address this issue, we describe a polynomial-time approximation that is within (1 -- 1/e) of the optimum by exploiting the submodularity of our criterion. This algorithm is extended to handle local structure in the GP, yielding significant speedups. We demonstrate the advantages of our approach on two real-world data sets.

Predicting probability distributions for surf height using an ensemble of mixture density networks

Unified activity management: supporting people in e-business

Meshing formal business processes with informal human collaborations is needed to support business activities.

INFER: a relational query language without the complexity of SQL.

Roles and relationships for unified activity management

This paper reports on three ethnographic studies of how people coordinate their activities in various work settings. The findings reported here are a derived set of relationships reflecting the nature of involvement of people in their activities. These findings were then tested by six analysts, who were conducting field studies of patterns of complex business activities. They used the derived relationships in the analysis of their data and in the representation of activity patterns. These usage cases revealed confusion between involvement relationships and job roles. Finally, several implications of these studies for designing an activity management prototype are presented.

VMSTAR: synthesizing scalable runtime environments for sensor networks

Intelligent light control using sensor networks

Scalable component abstractions

We identify three programming language abstractions for the construction of re-usable components: abstract type members, explicit selftypes and symmetric mixin composition. Together, these abstractions enable us to transform an arbitrary assembly of static program parts with hard references between them into a system of re-usable components. The transformation maintains the structure of the original system. We demonstrate this approach in two case studies, a subject/observer framework and a compiler front-end.

Optimal channel rate allocation for multimedia communication over fading wireless channels

What is worth learning from parallel workloads?

Learning useful and predictable features from past workloads and exploiting them well is a major source of improvement in many operating system problems. We review known parallel workload features, and argue that the correct approach for future on-line algorithm design as well as workload modeling is user- and session-based modeling, instead of analyzing jobs directly as done today. We then provide statistically sound answers to two basic questions: Which user and session features are central enough to be potentially useful, answered using Principal Component Analysis, and which user and session classes exist and how they can be identified on-line, answered using K-means clustering. We identify variable sets that explain over 80% of the variance between sessions and between users, and also identify five stable session classes (clusters) and four stable user classes. Our analysis is based on logs from seven different parallel supercomputers, spanning over 87 months, which are analyzed together to ensure that results are location- and architecture-neutral.

Agent-based trust model involving multiple qualities

A key limitation of current Web services standards is the inability to differentiate service instances at runtime using (nonfunctional) qualities of services (QoS). Such differentiation is necessary to allow for runtime selection and binding to service instances in a manner that continually adapts selected services to the service consumer’s preferences and needs. Quality values are volatile, depend on the trust associated with instances and service providers, and also depend on the needs of service consumers. We propose a multiagent framework where agents consider the consumers ’ QoS preferences, determine trust levels to associate to service instances and providers, and automatically select service instances on a consumer’s behalf. The service agents use a trust model that is centered on a shared conceptualization for QoS (ontology) and a QoS preference model that considers consumer’s tradeoffs among qualities as well as relationships between qualities. We evaluate our approach via simulations on simplified but realistic service instances and service consumers. Our results show that using these considerations for QoS, service agents are able to determine over time the ‘best ’ service selection for a consumer.

An algebraic decision diagram (ADD) based technique to find leakage histograms of combinational designs

Self-organizing distributed collaborative filtering

Bootstrapping dictionaries for cross-language information retrieval

Context-sensitive program analysis as database queries

The pauseless GC algorithm

Crawling a country

Beyond "from" and "received"

Assessing differential usage of usenet social accounting meta-data

e- e- probes of the littlest Higgs

Mapping A Communication Tree With Correlation Of Packets

Information fusion and uncertainty management for biological multisensor systems

Update on the INCITS W1.1 standard for evaluating the color rendition of printing systems

The color rendition ad hoc team of INCITS W1.1 is working to address issues related to color and tone reproduction for printed output and its perceptual impact on color image quality. The scope of the work includes accuracy of specified colors with an emphasis on memory colors, color gamut, and the effective use of tone levels, including issues related to contouring. The team has identified three sub-attributes of color rendition: 1) color quantization, defined as the ability to merge colors where needed; 2) color scale, defined as the ability to distinguish color where needed; and 3) color fidelity, defined as a balance of colorimetric accuracy, in cases where a reference exists, and pleasing overall color appearance. Visual definitions and descriptions of how these sub-attributes are perceived have been developed. The team is presently working to define measurement methods for the sub-attributes, with the focus in 2004 being on color fidelity. This presentation will review the definitions and appearance of the proposed sub-attributes and the progress toward developing test targets and associated measurement methods to quantify the color quantization sub-attribute. The remainder of the discussion will focus on the recent progress made in developing measurement methods for the color fidelity sub-attribute.

Wi-fi walkman: a wireless handheld that shares and recommends music on peer-to-peer networks

Feedback and Feedforward Control for Electrohydraulic Valve Systems Using a Novel System Identification Technique

Picturing Usenet: Mapping Computer-Mediated Collective Action

SAMPLE: Statistical Network Link Modelling in an On-Demand Probabilistic Routing Protocol for Ad Hoc Networks

Using feedback in collaborative reinforcement learning to adaptively optimize MANET routing

A Process-Complete Automatic Acceptance Testing Framework

We present a new automated software acceptance tests framework. The framework is novel in supporting the entire lifecycle and all QA activities, including test maintenance over multiple versions, interaction with programmers and business analysts, traceability to specifications, multi-user test cases and more. This enables a significant increase in QA productivity and product quality. We compare our framework to other available tools, products and frameworks, and present several patterns and anti-patterns for implementing a successful automated acceptance testing solution.

Distributed query processing using partitioned inverted files

Toward Web services interaction styles

Service-oriented architectures (SOAs) are fundamentally changing the way in which we conceptualize and design business applications. A SOA-based application typically composes various distributed functions, including some possibly provided by external parties such as independent businesses. The key advantage of SOAs is the resulting dynamism, since the composed parts can be readily swapped out in favor of others of like functionality. SOA environments thus reflect the dynamism of human socioeconomic environments where businesses interact, collaborate, and expose services to each other in order to jointly create value. This paper presents a multiagent model for Web services and catalogs architectural styles that are key for SOA applications. It conceptually evaluates the styles by showing the kinds of service usages and the resulting dynamic interactions that they enable.

Leveraging Social Networks to Fight Spam

Mobility and routing management for heterogeneous multi-hop wireless networks

XML retrieval

Test-driven development as a defect-reduction practice

Test-driven development is a software development practice that has been used sporadically for decades. With this practice, test cases (preferably automated) are incrementally written before production code is implemented. Test-driven development has recently re-emerged as a critical enabling practice of the extreme programming software development methodology. We ran a case study of this practice at IBM. In the process, a thorough suite of automated test cases was produced after UML design. In this case study, we found that the code developed using a test-driven development practice showed, during functional verification and regression tests, approximately 40% fewer defects than a baseline prior product developed in a more traditional fashion. The productivity of the team was not impacted by the additional focus on producing automated test cases. This test suite aids in future enhancements and maintenance of this code. The case study and the results are discussed in detail.

Approximate Distributed Kalman Filtering in Sensor Networks with Quantifiable Performance

Improving and Stabilizing Parallel Computer Performance Using Adaptive Backfilling

The scheduler is a key component in determining the overall performance of a parallel computer, and as we show here, the schedulers in wide use today exhibit large unexplained gaps in performance during their operation. Also, different scheduling algorithms often vary in the gaps they show, suggesting that choosing the correct scheduler for each time frame can improve overall performance. We present two adaptive algorithms that achieve this: One chooses by recent past performance, and the other by the recent average degree of parallelism, which is shown to be correlated to algorithmic superiority. Simulation results for the algorithms on production workloads are analyzed, and illustrate unique features of the chaotic temporal structure of parallel workloads. We provide best parameter configurations for each algorithm, which both achieve average improvements of 10% in performance and 35% in stability for the tested workloads.

Optical flame detection using large-scale artificial neural networks

A model for intelligent hydrocarbon flame detection using artificial neural networks (ANN) with a large number of inputs is presented. Joint time-frequency analysis in the form of short-time Fourier transform was used for extracting the relevant features from infrared sensor signals. After appropriate scaling, this information was provided as an input for the ANN training algorithm based on conjugate-gradient (CG) descent method. A classification scheme with trained ANN connection weights was implemented on a digital signal processor for an industrial hydrocarbon flame detector.

Online bagging and boosting

Real-time video delivery using peer-to-peer bartering networks and multiple description coding

Heuristics for Profile-Driven Method-Level Speculative Parallelization

Intent inference using a potential field model of environmental influences

Bias Analysis in Text Classification for Highly Skewed Data

Data Triage: an adaptive architecture for load shedding in TelegraphCQ

Mobile ad hoc network security - a taxonomy

Self-Adjusting Trust and Selection for Web Services

Service-oriented architectures enable services to be dynamically selected and integrated at runtime, thus enabling system flexibility and adaptiveness - autonomic attributes that are key for modern business needs. However, current techniques provide no support for actually making rational selections, which are key to accomplishing autonomic behavior. We develop a multiagent framework based on an ontology for QoS and a new model of trust. The agents form an ecosystem in which they help each other select services best matching their preferences. We evaluate by simulation the agents' effectiveness in maintaining self-adjusting trust

Combining Visualization and Statistical Analysis to Improve Operator Confidence and Efficiency for Failure Detection and Localization

Achieving stability in networks of input-queued switches using a local online scheduling policy