IBM

About: IBM is a company organization based out in Armonk, New York, United States. It is known for research contribution in the topics: Layer (electronics) & Signal. The organization has 134567 authors who have published 253905 publications receiving 7458795 citations. The organization is also known as: International Business Machines Corporation & Big Blue.

Direct Importance Estimation with Model Selection and Its Application to Covariate Shift Adaptation

Abstract: A situation where training and test samples follow different input distributions is called covariate shift. Under covariate shift, standard learning methods such as maximum likelihood estimation are no longer consistent—weighted variants according to the ratio of test and training input densities are consistent. Therefore, accurately estimating the density ratio, called the importance, is one of the key issues in covariate shift adaptation. A naive approach to this task is to first estimate training and test input densities separately and then estimate the importance by taking the ratio of the estimated densities. However, this naive approach tends to perform poorly since density estimation is a hard task particularly in high dimensional cases. In this paper, we propose a direct importance estimation method that does not involve density estimation. Our method is equipped with a natural cross validation procedure and hence tuning parameters such as the kernel width can be objectively optimized. Simulations illustrate the usefulness of our approach.

Randomized Algorithms

Abstract: For many applications, a randomized algorithm is either the simplest or the fastest algorithm available, and sometimes both. This book introduces the basic concepts in the design and analysis of randomized algorithms. The first part of the text presents basic tools such as probability theory and probabilistic analysis that are frequently used in algorithmic applications. Algorithmic examples are also given to illustrate the use of each tool in a concrete setting. In the second part of the book, each chapter focuses on an important area to which randomized algorithms can be applied, providing a comprehensive and representative selection of the algorithms that might be used in each of these areas. Although written primarily as a text for advanced undergraduates and graduate students, this book should also prove invaluable as a reference for professionals and researchers.

Quantum-mechanical modeling of electron tunneling current from the inversion layer of ultra-thin-oxide nMOSFET's

Abstract: Quantum-mechanical modeling of electron tunneling current from the quantized inversion layer of ultra-thin-oxide (<40 /spl Aring/) nMOSFET's is presented, together with experimental verification. An accurate determination of the physical oxide thickness is achieved by fitting experimentally measured capacitance-versus-voltage curves to quantum-mechanically simulated capacitance-versus-voltage results. The lifetimes of quasibound states and the direct tunneling current are calculated using a transverse-resonant method. These results are used to project an oxide scaling limit of 20 /spl Aring/ before the chip standby power becomes excessive due to tunneling currents,.

Systematic study of insulator-metal transitions in perovskites RNiO3 (R=Pr,Nd,Sm,Eu) due to closing of charge-transfer gap.

Abstract: The detailed behavior of the phase transitions was mapped out for the series R${\mathrm{NiO}}_{3}$ as a function of the rare earth (R). A sharp insulator-metal transition is observed, which depends strongly on R.Forsmall$R it occurs at a higher temperature than the antiferromagnetic ordering (measured by muon-spin relaxation). By increasing either the temperature or the size of R, an insulator-metal transition is observed, most probably caused by the closing of the charge-transfer gap, induced by increase in the electronic bandwidth.

Mining process models from workflow logs

Abstract: Modern enterprises increasingly use the workflow paradigm to prescribe how business processes should be performed. Processes are typically modeled as annotated activity graphs. We present an approach for a system that constructs process models from logs of past, unstructured executions of the given process. The graph so produced conforms to the dependencies and put executions present in the log. By providing models that capture the previous executions of the process, this technique allows easier introduction of a workflow system and evaluation and evolution of existing process models. We also present results from applying the algorithm to synthetic data sets as well as process logs obtained from an IBM Flowmark installation.