A5086856121- Sparse and Compressive Sensing Techniques
- Statistical Methods and Inference
- Direction-of-Arrival Estimation Techniques
- Distributed Sensor Networks and Detection Algorithms
- Control Systems and Identification
- Risk and Portfolio Optimization
- Advanced Statistical Methods and Models
- Mathematical Inequalities and Applications
- Tensor decomposition and applications
- Advanced Neuroimaging Techniques and Applications
- Computer Graphics and Visualization Techniques
- Ancient Mediterranean Archaeology and History
- History and advancements in chemistry
- Face and Expression Recognition
- Ancient and Medieval Archaeology Studies
- Chemistry and Stereochemistry Studies
- Analytical Chemistry and Chromatography
- Blind Source Separation Techniques
- Reservoir Engineering and Simulation Methods
- Archaeology and Historical Studies
ETH Zurich
2007-2019
University of Zurich
2011
Paul Scherrer Institute
2007
Large scale and structurally complex volume datasets from high-resolution 3D imaging devices or computational simulations pose a number of technical challenges for interactive visual analysis. In this paper, we present the first integration multiscale representation based on tensor approximation within GPU-accelerated out-of-core multiresolution rendering framework. Specific contributions include (a) hierarchical brick-tensor decomposition approach pre-processing large data, (b) GPU...
Many results have been proved for various nuclear norm penalized estimators of the uniform sampling matrix completion problem. However, most these are not robust: in cases quadratic loss function and its modifications used. We consider robust using two well-known functions: absolute value Huber loss. Under several conditions on sparsity problem (i.e., rank parameter matrix) regularity risk sharp nonsharp oracle inequalities shown to hold with high probability. As a consequence, asymptotic...
Supervised learning methods with missing data have been extensively studied not just due to the techniques related low‐rank matrix completion. Also, in unsupervised learning, one often relies on imputation methods. As a matter of fact, values induce bias various estimators such as sample covariance matrix. In present paper, convex method for sparse subspace estimation is extended case and corrupted measurements. This done by correcting instead imputing values. The estimator then used an...
Many statistical estimation procedures lead to nonconvex optimization problems. Algorithms solve these problems are often guaranteed output a stationary point of the problem. Oracle inequalities an important theoretical instrument assess performance estimator. results have focused on properties uncomputable (global) minimum or maximum. In this paper, general framework used for convex derive oracle points is extended. A main new ingredient that they sharp: show closeness best approximation...
In this work we describe a highly automated procedure ('workflow') for the analysis of electronic and molecular structure data obtained from quantum chemical computations. The generated as part workflow are archived in an XML/CML database. These processed by means statistical analysis. This production machinery is applied towards interference dependencies between electron delocalization properties functionalized linearly ?-conjugated compounds. information source generation rules or...
Many results have been proved for various nuclear norm penalized estimators of the uniform sampling matrix completion problem. However, most these are not robust: in cases quadratic loss function and its modifications used. We consider robust using two well-known functions: absolute value Huber loss. Under several conditions on sparsity problem (i.e. rank parameter matrix) regularity risk sharp non-sharp oracle inequalities shown to hold with high probability. As a consequence, asymptotic...
Supervised learning methods with missing data have been extensively studied not just due to the techniques related low-rank matrix completion. Also in unsupervised one often relies on imputation methods. As a matter of fact, values induce bias various estimators such as sample covariance matrix. In present paper, convex method for sparse subspace estimation is extended case and corrupted measurements. This done by correcting instead imputing values. The estimator then used an initial value...
Many statistical estimation procedures lead to nonconvex optimization problems. Algorithms solve these are often guaranteed output a stationary point of the problem. Oracle inequalities an important theoretical instrument asses performance estimator. results have focused on properties uncomputable (global) minimum or maximum. In present work general framework used for convex problems derive oracle points is extended. A main new ingredient that they sharp: show closeness best approximation...