We introduce the concept of Hypoelliptic Diffusion Maps (HDM), a framework generalizing Diffusion Maps in the context of manifold learning and dimensionality reduction. Standard non-linear dimensionality reduction methods (e.g., LLE, ISOMAP, Laplacian Eigenmaps, Diffusion Maps) focus on mining massive data sets using weighted affinity graphs; Orientable Diffusion Maps and Vector Diffusion Maps enrich these graphs by attaching to each node also some local geometry. HDM likewise considers a scenario where each node possesses additional structure, which is now itself of interest to investigate. Virtually, HDM augments the original data set with attached structures, and provides tools for studying and organizing the augmented ensemble. The goal is to obtain information on individual structures attached to the nodes and on the relationship between structures attached to nearby nodes, so as to study the underlying manifold from which the nodes are sampled. In this paper, we analyze HDM on tangent bundles, revealing its intimate connection with sub-Riemannian geometry and a family of hypoelliptic differential operators. In a later paper, we shall consider more general fibre bundles.

We present an integrated approach to structure and parameter estimation in latent tree graphical models, where some nodes are hidden. Our overall approach follows a "divide-and-conquer" strategy that learns models over small groups of variables and iteratively merges into a global solution. The structure learning involves combinatorial operations such as minimum spanning tree construction and local recursive grouping; the parameter learning is based on the method of moments and on tensor decompositions. Our method is guaranteed to correctly recover the unknown tree structure and the model parameters with low sample complexity for the class of linear multivariate latent tree models which includes discrete and Gaussian distributions, and Gaussian mixtures. Our bulk asynchronous parallel algorithm is implemented in parallel using the OpenMP framework and scales logarithmically with the number of variables and linearly with dimensionality of each variable. Our experiments confirm a high degree of efficiency and accuracy on large datasets of electronic health records. The proposed algorithm also generates intuitive and clinically meaningful disease hierarchies.

We characterize the performance of sequential information guided sensing, Info-Greedy Sensing, when there is a mismatch between the true signal model and the assumed model, which may be a sample estimate. In particular, we consider a setup where the signal is low-rank Gaussian and the measurements are taken in the directions of eigenvectors of the covariance matrix in a decreasing order of eigenvalues. We establish a set of performance bounds when a mismatched covariance matrix is used, in terms of the gap of signal posterior entropy, as well as the additional amount of power required to achieve the same signal recovery precision. Based on this, we further study how to choose an initialization for Info-Greedy Sensing using the sample covariance matrix, or using an efficient covariance sketching scheme.

This paper describes an end-to-end learning framework that allows a novice to create a model from data easily by helping structure the model building process and capturing extended aspects of domain knowledge. By treating the whole modeling process interactively and exploiting high-level knowledge in the form of an ontology, the framework is able to aid the user in a number of ways, including in helping to avoid pitfalls such as data dredging. Prudence must be exercised to avoid these hazards: certain conclusions may be supported by extra knowledge if, for example, there are reasons to trust a particular narrower set of hypotheses. This paper adopts the solution of using higher-level knowledge in order to allow this sort of domain knowledge to be inferred automatically, thereby selecting only relevant input attributes and thence constraining the hypothesis space. We describe how the framework automatically exploits structured knowledge in an ontology to identify relevant concepts, and how a data extraction component can make use of online data sources to find measurements of those concepts so that their relevance can be evaluated. To validate our approach, models of four different problem domains were built using our implementation of the framework. Prediction error on unseen examples of these models show that our framework, making use of the ontology, helps to improve model generalization.

Named Entity Recognition (NER) and Named Entity Normalization (NEN) refer to the recognition and normalization of raw texts to known entities. From the perspective of recruitment innovation, professional skill characterization and normalization render human capital data more meaningful both commercially and socially. Accurate and detailed normalization of skills is the key for the predictive analysis of labor market dynamics. Such analytics help bridge the skills gap between employers and candidate workers by matching the right talent for the right job and identifying in-demand skills for workforce training programs. This can also work towards the social goal of providing more job opportunities to the community. In this paper we propose an automated approach for skill entity recognition and optimal normalization. The proposed system has two components: 1) Skills taxonomy generation, which employs vocational skill related sections of resumes and Wikipedia categories to define and develop a taxonomy of professional skills; 2) Skills tagging, which leverages properties of semantic word vectors to recognize and normalize relevant skills in input text. By sampling based end-user evaluation, the current system attains 91% accuracy on the taxonomy generation and 82% accuracy on the skills tagging tasks. The beta version of the system is currently applied in various big data and business intelligence applications for workforce analytics and career track projections at CareerBuilder.

Complex operational environments require improved tactical mission command capabilities with a high level of interoperability among coalition control and command (C2) systems. This paper focuses on two areas of interest: decision support based on automated planning and Service Oriented Architecture (SOA) for rapid service development. Previous experiments were performed bilaterally by US, France and Germany to focus on collaborative mission planning using Web Services (WSs). The results reported herein were obtained from a unified experiment performed by US, France and Germany involving a common scenario. The operational benefit from the experimentation has been to improve mutual understanding among allied forces, to dynamically plan for assistance among ground support troops (logistics, MEDE- VAC, and other areas) as well as to improve their coordination. The effort addressed system design, and integration within an experimental framework. It enabled the evolution of the CERDEC Mission Command Gateway (MCG) architecture as well as a constraint based planner ”ORTAC”, developed by French DGA and Sagem. It takes into account near real-time multimodal Situation Awareness and readiness status from tactical edge units. The trilateral experiment, entitled From Data to Decision included Net-Centric manned and unmanned assets from all three nations (France - Germany - US) operating as a cohesive coalition force while preserving command and support relationships as required through their respective chains of command.

This paper addresses the problem of named entity recognition (NER) in travel-related search queries. NER is an important step toward a richer understanding of user-generated inputs in information retrieval systems. NER in queries is challenging due to minimal context and few structural clues. NER in restricted-domain queries is useful in vertical search applications, for example following query classification in general search. This paper describes an efficient machine learning-based solution for the high-quality extraction of semantic entities from query inputs in a restricted-domain information retrieval setting. We apply a conditional random field (CRF) sequence model to travel-domain search queries and achieve high-accuracy results. Our approach yields an overall F1 score of 86.4% on a held-out test set, outperforming a baseline score of 82.0% on a CRF with standard features. The resulting NER classifier is currently in use in a real-life travel search engine.

We developed a tool to solve a problem of position assignment within the IT Ford College Graduate program. This position assignment tool was first developed in 2012 and has been used successfully since then. The tool has since evolved for use with several other position assignment and related tasks with other similar programs in Ford Motor Company. This paper will describe the creation of this tool and how we have applied it, focusing on the need for developing such a tool, and how the continued development of this tool will benefit its users and the company.

This paper describes a novel combination of Java program analysis and automated learning and planning architecture to the domain of Java vulnerability analysis. The key feature of our "HACKAR:Helpful Advice for Code Knowledge and Attack Resilience'' system is its ability to analyze Java programs at development-time, identifying vulnerabilities and ways to avoid them. HACKAR uses an improved version of NASA's Java PathFinder (JPF) to execute Java programs and identify vulnerabilities. The system features new Hierarchical Task Network (HTN) learning algorithms that (1) advance state-of-the-art HTN learners with reasoning about numeric constraints, failures, and more general cases of recursion, and (2) contribute to problem-solving by learning a hierarchical dataflow representation of the program from the inputs of the program. Empirical evaluation demonstrates that HACKAR was able to suggest fixes for all of our test program suites. It also shows that HACKAR can analyze programs with string inputs that original JPF implementation cannot.

PDS (Process Diagnosis System) is an expert system shell developed in the early 1980's. It could handle thousands of sensor inputs and produce thousands of diagnostic messages with confidence factors based on complex logic designed to mimic the thinking of human experts. PDS went into commercial operation in 1985 to monitor seven power plant generators from a centralized diagnostic center at Westinghouse Power Generation headquarters. In the 1990’s the popularity of advanced technology gas turbines provided a renaissance in PDS utilization. The software has undergone rewrites and improvements since its inception, and the current PCPDS now supports the Siemens Power Diagnostics® Center with centralized rule based monitoring of over 1200 gas turbines, steam turbines, and generators.