Papers.

EXAMINING NCAA/NFL MARKET EFFICIENCY

Engagements: Building Eventually ACiD Business Transactions

EMuRgency – A Basic Concept for an AI Driven Volunteer Notification System for Integrating Laypersons into Emergency Medical Services

Efficient Algebraic Representations for Throughput-Oriented Algorithms

This dissertation explores the design and deployment of memory-efficient tensor product representations for throughput-oriented processors. These representations lead naturally to algorithms with excellent data-locality and can be used to construct efficient and scalable representations for hierarchical tensor-based algorithms.

DysWexia: Textos más Accesibles para Personas con Dislexia

DysWebxia: Textos m as Accesibles para Personas con Dislexia DysWebxia: Making Texts More Accessible for People with Dyslexia

Distributional Semantics for Answer Re-ranking in Question Answering

Abstract. This paper investigates the role of Distributional Semantic Models (DSMs) into a Question Answering (QA) system. Our purpose is to exploit DSMs for answer re-ranking in QuestionCube, a framework for building QA systems. DSMs model words as points in a geometric space, also known as semantic space. Words are similar if they are close in that space. Our idea is that DSMs approaches can help to compute relatedness between users ’ questions and candidate answers by exploiting paradigmatic relations between words, thus providing better answer reranking. Results of the evaluation, carried out on the CLEF2010 QA dataset, prove the effectiveness of the proposed approach. 1

Design Patterns in Dynamic Programming

Design and Analysis of Multi-Core Architecture for Cyber-Physical Systems

Decentralized Entity-Level Modeling for Coreference Resolution

Data Insights: New Ways to Visualize and Make Sense of Data

Crude oil prices: speculation versus fundamentals

CMUML System for KBP 2013 Slot Filling

Chorus: a crowd-powered conversational assistant

Big Data or Right Data

Better Word Representations with Recursive Neural Networks for Morphology

Basis Technology at TAC 2013 Entity Linking

Articulatory features for large vocabulary speech recognition

Analyzing microtext : papers from the AAAI Spring Symposium

Analysis of the user queries of an e-commerce bookstore in terms of the Library of Congress classification and key publishers

ADAM: Genomics Formats and Processing Patterns for Cloud Scale Computing

Current genomics data formats and processing pipelines are not designed to scale well to large datasets. The current Sequence/Binary Alignment/Map (SAM/BAM) formats were intended for single node processing. There have been attempts to adapt BAM to distributed computing environments, but they see limited scalability past eight nodes. Additionally, due to the lack of an explicit data schema, there are well known incompatibilities between libraries that implement SAM/BAM/Variant Call Format (VCF) data access. To address these problems, we introduce ADAM, a set of formats, APIs, and processing stage implementations for genomic data. ADAM is fully open source under the Apache 2 license, and is implemented on top of Avro and Parquet for data storage.

A network science perspective of a distributed reputation mechanism

A Cross-language Study on Automatic Speech Disfluency Detection

A Case for Ordinal Peer-evaluation in MOOCs

A Scalable Architecture For Web Deployment of Spoken Dialogue Systems

Twenty Years of Language Resource Development and Distribution: A Progress Report on LDC Activities

Larry A. Greene and Anke Ortlepp, eds., Germans and African Americans: Two Centuries of Exchange. Jackson: University Press of Mississippi, 2011. Pp. 304. Cloth $55.00.

The Impact of Programming Languages in Code Cloning

Global Analysis of Security and Trust Perceptions in Web Design for E-Commerce

Probabilistically Bounded Staleness for Practical Partial Quorums

An Introduction to Conditional Random Fields

Facts, trends and challenges in modern software development

The IT industry is not new to change and evolution, however, we are now in an era of two fundamental waves of IT changes. First, the post-PC era, where mobile devices and tablet-like devices are giving end-users the ability to consume information when they want it and where they want it. Second, the post-server era where companies no longer need to neither buy nor provision servers in their own data centres but instead rent the compute resources as needed. This twin change has direct consequences to how end-users consume software, how that software is produced, and how it is delivered.

Avatara: OLAP for Web-scale Analytics Products

Probabilistically bounded staleness for practical partial quorums

Distributed GraphLab: A Framework for Machine Learning in the Cloud

While high-level data parallel frameworks, like MapReduce, simplify the design and implementation of large-scale data processing systems, they do not naturally or efficiently support many important data mining and machine learning algorithms and can lead to inefficient learning systems. To help fill this critical void, we introduced the GraphLab abstraction which naturally expresses asynchronous, dynamic, graph-parallel computation while ensuring data consistency and achieving a high degree of parallel performance in the shared-memory setting. In this paper, we extend the GraphLab framework to the substantially more challenging distributed setting while preserving strong data consistency guarantees. We develop graph based extensions to pipelined locking and data versioning to reduce network congestion and mitigate the effect of network latency. We also introduce fault tolerance to the GraphLab abstraction using the classic Chandy-Lamport snapshot algorithm and demonstrate how it can be easily implemented by exploiting the GraphLab abstraction itself. Finally, we evaluate our distributed implementation of the GraphLab abstraction on a large Amazon EC2 deployment and show 1-2 orders of magnitude performance gains over Hadoop-based implementations.

IP-FCM Measures Physiologic Protein-Protein Interactions Modulated by Signal Transduction and Small-Molecule Drug Inhibition

Protein-protein interactions (PPI) mediate the formation of intermolecular networks that control biological signaling. For this reason, PPIs are of outstanding interest in pharmacology, as they display high specificity and may represent a vast pool of potentially druggable targets. However, the study of physiologic PPIs can be limited by conventional assays that often have large sample requirements and relatively low sensitivity. Here, we build on a novel method, immunoprecipitation detected by flow cytometry (IP-FCM), to assess PPI modulation during either signal transduction or pharmacologic inhibition by two different classes of small-molecule compounds. First, we showed that IP-FCM can detect statistically significant differences in samples possessing a defined PPI change as low as 10%. This sensitivity allowed IP-FCM to detect a PPI that increases transiently during T cell signaling, the antigen-inducible interaction between ZAP70 and the T cell antigen receptor (TCR)/CD3 complex. In contrast, IP-FCM detected no ZAP70 recruitment when T cells were stimulated with antigen in the presence of the src-family kinase inhibitor, PP2. Further, we tested whether IP-FCM possessed sufficient sensitivity to detect the effect of a second, rare class of compounds called SMIPPI (small-molecule inhibitor of PPI). We found that the first-generation non-optimized SMIPPI, Ro-26-4550, inhibited the IL-2:CD25 interaction detected by IP-FCM. This inhibition was detectable using either a recombinant CD25-Fc chimera or physiologic full-length CD25 captured from T cell lysates. Thus, we demonstrate that IP-FCM is a sensitive tool for measuring physiologic PPIs that are modulated by signal transduction and pharmacologic inhibition.

Sequencing three crocodilian genomes to illuminate the evolution of archosaurs and amniotes

Motion planning under uncertainty using iterative local optimization in belief space

We present a new approach to motion planning under sensing and motion uncertainty by computing a locally optimal solution to a continuous partially observable Markov decision process (POMDP). Our approach represents beliefs (the distributions of the robot’s state estimate) by Gaussian distributions and is applicable to robot systems with non-linear dynamics and observation models. The method follows the general POMDP solution framework in which we approximate the belief dynamics using an extended Kalman filter and represent the value function by a quadratic function that is valid in the vicinity of a nominal trajectory through belief space. Using a belief space variant of iterative LQG (iLQG), our approach iterates with second-order convergence towards a linear control policy over the belief space that is locally optimal with respect to a user-defined cost function. Unlike previous work, our approach does not assume maximum-likelihood observations, does not assume fixed estimator or control gains, takes into account obstacles in the environment, and does not require discretization of the state and action spaces. The running time of the algorithm is polynomial (O[n 6 ]) in the dimension n of the state space. We demonstrate the potential of our approach in simulation for holonomic and non-holonomic robots maneuvering through environments with obstacles with noisy and partial sensing and with non-linear dynamics and observation models.

Uncovering the role of genomic “dark matter” in human disease

Session details: AdMIRe'12 workshop 6

Session details: European track presentations

Session details: Industry track presentations

Shark: SQL and rich analytics at scale

Shark is a new data analysis system that marries query processing with complex analytics on large clusters. It leverages a novel distributed memory abstraction to provide a unified engine that can run SQL queries and sophisticated analytics functions (e.g. iterative machine learning) at scale, and efficiently recovers from failures mid-query. This allows Shark to run SQL queries up to 100X faster than Apache Hive, and machine learning programs more than 100X faster than Hadoop. Unlike previous systems, Shark shows that it is possible to achieve these speedups while retaining a MapReduce-like execution engine, and the fine-grained fault tolerance properties that such engine provides. It extends such an engine in several ways, including column-oriented in-memory storage and dynamic mid-query replanning, to effectively execute SQL. The result is a system that matches the speedups reported for MPP analytic databases over MapReduce, while offering fault tolerance properties and complex analytics capabilities that they lack.

ECIR 2012

Tradeoffs in CDN Designs for Throughput Oriented Traffic

CrowdTiles: presenting crowd-based information for event-driven information needs

User engagement

The potential dangers of causal consistency and an explicit solution

Crowdsourced user interface testing for multimedia applications

Condos and Clouds

Uses the condominium analogy to explain the constraints and standardization that cloud platforms impose on applications in exchange for manageability at scale.