In this paper we present an algorithm for the design of robust model-based predictive control for hybrid system under uncertainty via parametric programming. The proposed min-max hybrid control scheme guarantees feasible plant operation for the maximum violation of uncertainty scenario. The key advantage of the proposed hybrid controller design is reduction in expensive, repetitive in nature on...

We consider statistical procedures for feature selection defined by a family of regularization problems with convex piecewise linear loss functions and penalties of l1 nature. Many known statistical procedures (e.g. quantile regression and support vector machines with l1 norm penalty) are subsumed under this category. Computationally, the regularization problems are linear programming (LP) prob...

Mean-Shift (MS) is a powerful non-parametric clustering method. Although good accuracy can be achieved, its computational cost is particularly expensive even on moderate data sets. In this paper, for the purpose of algorithm speedup, we develop an agglomerative MS clustering method called Agglo-MS, along with its mode-seeking ability and convergence property analysis. Our method is built upon a...

This review summarizes recent attempts to reconstruct the expansion history of the Universe and to probe the nature of dark energy. Reconstruction methods can be broadly classified into parametric and non-parametric approaches. It is encouraging that, even with the limited observational data currently available, different approaches give consistent results for the reconstruction of the Hubble p...

In this talk, I present some theoretical and empirical properties of the uniform and biased-bootstrap for generalized method of moments (GMM) models. The version of the biased-bootstrap used in this paper is a form of weighted bootstrap with weights chosen to satisfy some constraints imposed by the model. A typical biased-bootstrap resample is obtained by resampling from a member within a pseud...

Modern Very Large Scale Integration (VLSI) designs require a tradeoff between cost efficiency and performance (circuit speed). Furthermore, the Design Space Exploration (DSE) of the cost-performance tradeoffs for the multi objective VLSI designs should also be fast and efficient in nature. This paper presents a novel accelerated DSE approach for the exploration of cost-performance tradeoffs of ...

This paper examines the relation between dividend policy, managerial ownership and debt-financing for a large sample of firms listed on NYSE, AMEX and NASDAQ. In addition to standard parametric estimation methods,weusea semi-parametric approach,whichhelps capturemoreeffectivelynon-linearities in thedata. In linewith the alignment effect ofmanagerial ownership, our results support a negative rel...

We introduce a non-parametric method for segmentation in regimeswitching time-series models. The approach is based on spectral clustering of target-regressor tuples and derives a switching regression tree, where regime switches are modeled by oblique splits. Such models can be learned efficiently from data, where clustering is used to propose one single split candidate at each split level. We u...

The implementation of a recently suggested class of structural change tests, which test for parameter instability in general parametric models, in the R language for statistical computing is described: Focus is given to the question how the conceptual tools can be translated into computational tools that reflect the properties and flexiblity of the underlying econometric metholody while being n...

We present a framework for pose-invariant face recognition using parametric linear subspace models as stored representations of known individuals. Each model can be t to an input, resulting in faces of known people whose head pose is aligned to the input face. The model's continuous nature enables the pose alignment to be very accurate, improving recognition performance, while its generalizatio...