Multi-dimensional sparse structured signal approximation using split Bregman iterations     Sparse approximation  Regularization  Fused-LASSO  Split Bregman  Multidimensional signals       font style='font-size:12px;'> 2013/5/2

The paper focuses on the sparse approximation of signals using overcomplete representations, such that it preserves the (prior) structure of multi-dimensional signals. The underlying optimization prob...

存档文本 存档附件 原文地址 文献传递

Local Gaussian process approximation for large computer experiments     sequential design  sequential updating  active learning  surrogate model  emulator  compactly supported covariance  local kriging neighborhoods       font style='font-size:12px;'> 2013/4/27

We provide a new approach to approximate emulation of large computer experiments. By focusing expressly on desirable properties of the predictive equations, we derive a family of local sequential desi...

存档文本 存档附件 原文地址 文献传递

Distribution of the largest eigenvalue for real Wishart and Gaussian random matrices and a simple approximation for the Tracy-Widom distribution     Random Matrix Theory  characteristic roots  largest eigenvalue  Tracy-Widom Distribution  Wishart Matrices  Gaussian Orthogonal Ensemble       font style='font-size:12px;'> 2012/11/23

We derive the exact distribution of the largest eigenvalue for finite dimensions real Wishart matrices and for the Gaussian Orthogonal Ensemble (GOE). We compare the exact distribution with the Tracy-...

存档文本 存档附件 原文地址 文献传递

Uniform Stability of a Particle Approximation of the Optimal Filter Derivative     Hidden Markov Models  State-Space Models  Sequential Monte Carlo       font style='font-size:12px;'> 2011/7/5

Sequential Monte Carlo methods, also known as particle methods, are a widely used set of computational tools for inference in non-linear non-Gaussian state-space models.

存档文本 存档附件 原文地址 文献传递