Kush R. Varshney

Sparse Representation in Structured Dictionaries With Application to Synthetic Aperture Radar

Sparse signal representations and approximations from overcomplete dictionaries have become an invaluable tool recently. In this paper, we develop a new, heuristic, graph-structured, sparse signal representation algorithm for overcomplete dictionaries that can be decomposed into subdictionaries and whose dictionary elements can be arranged in a hierarchy. Around this algorithm, we construct a methodology for advanced image formation in wide-angle synthetic aperture radar (SAR), defining an approach for joint anisotropy characterization and image formation. Additionally, we develop a coordinate descent method for jointly optimizing a parameterized dictionary and recovering a sparse representation using that dictionary. The motivation is to characterize a phenomenon in wide-angle SAR that has not been given much attention before: migratory scattering centers, i.e., scatterers whose apparent spatial location depends on aspect angle. Finally, we address the topic of recovering solutions that are sparse in more than one objective domain by introducing a suitable sparsifying cost function. We encode geometric objectives into SAR image formation through sparsity in two domains, including the normal parameter space of the Hough transform.

Sparsity-Driven Synthetic Aperture Radar Imaging: Reconstruction, autofocusing, moving targets, and compressed sensing

This article presents a survey of recent research on sparsity-driven synthetic aperture radar (SAR) imaging. In particular, it reviews 1) the analysis and synthesis-based sparse signal representation formulations for SAR image formation together with the associated imaging results, 2) sparsity-based methods for wide-angle SAR imaging and anisotropy characterization, 3) sparsity-based methods for joint imaging and autofocusing from data with phase errors, 4) techniques for exploiting sparsity for SAR imaging of scenes containing moving objects, and 5) recent work on compressed sensing (CS)-based analysis and design of SAR sensing missions.

Quantization of Prior Probabilities for Hypothesis Testing

In this paper, Bayesian hypothesis testing is investigated when the prior probabilities of the hypotheses, taken as a random vector, are quantized. Nearest neighbor and centroid conditions are derived using mean Bayes risk error (MBRE) as a distortion measure for quantization. A high-resolution approximation to the distortion-rate function is also obtained. Human decision making in segregated populations is studied assuming Bayesian hypothesis testing with quantized priors.

Dynamic matrix factorization: A state space approach

Matrix factorization from a small number of observed entries has recently garnered much attention as the key ingredient of successful recommendation systems. One unresolved problem in this area is how to adapt current methods to handle changing user preferences over time. Recent proposals to address this issue are heuristic in nature and do not fully exploit the time-dependent structure of the problem. As a principled and general temporal formulation, we propose a dynamical state space model of matrix factorization. Our proposal builds upon probabilistic matrix factorization, a Bayesian model with Gaussian priors. We utilize results in state tracking, i.e. the Kalman filter, to provide accurate recommendations in the presence of both process and measurement noise. We show how system parameters can be learned via expectation-maximization and provide comparisons to current published techniques.

Cognition as a part of computational creativity

Computational creativity and cognitive computing are distinct fields that have developed in a parallel fashion. In this paper, we examine the relationship between the two, concluding that the two fields overlap in one precise way: the evaluation or assessment of artifacts with respect to creativity. Furthermore, we discuss a particular instance of computational creativity, culinary recipe design, and how cognitive informatics and cognitive computation enter into the domain.

Linear Dimensionality Reduction for Margin-Based Classification: High-Dimensional Data and Sensor Networks

Low-dimensional statistics of measurements play an important role in detection problems, including those encountered in sensor networks. In this work, we focus on learning low-dimensional linear statistics of high-dimensional measurement data along with decision rules defined in the low-dimensional space in the case when the probability density of the measurements and class labels is not given, but a training set of samples from this distribution is given. We pose a joint optimization problem for linear dimensionality reduction and margin-based classification, and develop a coordinate descent algorithm on the Stiefel manifold for its solution. Although the coordinate descent is not guaranteed to find the globally optimal solution, crucially, its alternating structure enables us to extend it for sensor networks with a message-passing approach requiring little communication. Linear dimensionality reduction prevents overfitting when learning from finite training data. In the sensor network setting, dimensionality reduction not only prevents overfitting, but also reduces power consumption due to communication. The learned reduced-dimensional space and decision rule is shown to be consistent and its Rademacher complexity is characterized. Experimental results are presented for a variety of datasets, including those from existing sensor networks, demonstrating the potential of our methodology in comparison with other dimensionality reduction approaches.

Predicting employee expertise for talent management in the enterprise

Strategic planning and talent management in large enterprises composed of knowledge workers requires complete, accurate, and up-to-date representation of the expertise of employees in a form that integrates with business processes. Like other similar organizations operating in dynamic environments, the IBM Corporation strives to maintain such current and correct information, specifically assessments of employees against job roles and skill sets from its expertise taxonomy. In this work, we deploy an analytics-driven solution that infers the expertise of employees through the mining of enterprise and social data that is not specifically generated and collected for expertise inference. We consider job role and specialty prediction and pose them as supervised classification problems. We evaluate a large number of feature sets, predictive models and postprocessing algorithms, and choose a combination for deployment. This expertise analytics system has been deployed for key employee population segments, yielding large reductions in manual effort and the ability to continually and consistently serve up-to-date and accurate data for several business functions. This expertise management system is in the process of being deployed throughout the corporation.

Collaborative Kalman Filtering for Dynamic Matrix Factorization

We propose a new algorithm for estimation, prediction, and recommendation named the collaborative Kalman filter. Suited for use in collaborative filtering settings encountered in recommendation systems with significant temporal dynamics in user preferences, the approach extends probabilistic matrix factorization in time through a state-space model. This leads to an estimation procedure with parallel Kalman filters and smoothers coupled through item factors. Learning of global parameters uses the expectation-maximization algorithm. The method is compared to existing techniques and performs favorably on both generated data and real-world movie recommendation data.

Business Analytics Based on Financial Time Series

Baniya merchants of the Mughal Empire, burgher merchants of the Swedish Empire, and chonin merchants of the Tokugawa Shogunate had the same questions on their mind as business people do today. To which townspeople should I sell my wares? Of folks that buy from me, are there any that might stop buying from me? Which groups buy which goods? Which saris should I show Ranna Devi to make as much money as I can? How much timber will people want in the coming weeks and months? The world has changed over the centuries with globalization, rapid transportation, instantaneous communication, expansive enterprises, and an explosion of data and signals along with ample computation to process them. In this new age, many continue to answer the aforementioned and other critical business questions in the old-fashioned way, i.e., based on intuition, gut instinct, and personal experience. In our globalized world, however, this is not sufficient anymore and it is essential to replace the business persons gut instinct with science. That science is business analytics. Business analytics is a broad umbrella entailing many problems and solutions, such as demand forecasting and conditioning, resource capacity planning, workforce planning, salesforce modeling and optimization, revenue forecasting, customer/product analytics, and enterprise recommender systems. In our department, we are in creasingly directing our focus on developing models and techniques to address such business problems. The goal of this article is to provide the reader with an overview of this interesting new area of research and then hone in on applications that might require the use of sophisticated signal processing methodologies and utilize financial signals as input.

Engineering safety in machine learning

Machine learning algorithms are increasingly influencing our decisions and interacting with us in all parts of our daily lives. Therefore, just like for power plants, highways, and myriad other engineered sociotechnical systems, we must consider the safety of systems involving machine learning. In this paper, we first discuss the definition of safety in terms of risk, epistemic uncertainty, and the harm incurred by unwanted outcomes. Then we examine dimensions, such as the choice of cost function and the appropriateness of minimizing the empirical average training cost, along which certain real-world applications may not be completely amenable to the foundational principle of modern statistical machine learning: empirical risk minimization. In particular, we note an emerging dichotomy of applications: ones in which safety is important and risk minimization is not the complete story (we name these Type A applications), and ones in which safety is not so critical and risk minimization is sufficient (we name these Type B applications). Finally, we discuss how four different strategies for achieving safety in engineering (inherently safe design, safety reserves, safe fail, and procedural safeguards) can be mapped to the machine learning context through inter-pretability and causality of predictive models, objectives beyond expected prediction accuracy, human involvement for labeling difficult or rare examples, and user experience design of software.