Groundwater permeability Easy to solve the forward problem: flow of groundwater given permeability of aquifer Inverse problem: determine permeability from.

Slides:



Advertisements
Similar presentations
Pattern Recognition and Machine Learning
Advertisements

Sampling Design, Spatial Allocation, and Proposed Analyses Don Stevens Department of Statistics Oregon State University.
Spatial point patterns and Geostatistics an introduction
EigenFaces and EigenPatches Useful model of variation in a region –Region must be fixed shape (eg rectangle) Developed for face recognition Generalised.
Pattern Recognition and Machine Learning
EDGE DETECTION.
STAT 497 APPLIED TIME SERIES ANALYSIS
Data mining and statistical learning - lecture 6
Gaussian process emulation of multiple outputs Tony O’Hagan, MUCM, Sheffield.
Meet the professor Friday, January 23 at SFU 4:30 Beer and snacks reception.
Midterm Review CS485/685 Computer Vision Prof. Bebis.
Nonstationary covariance structures II NRCSE. Drawbacks with deformation approach Oversmoothing of large areas Local deformations not local part of global.
Using wavelet tools to estimate and assess trends in atmospheric data Peter Guttorp University of Washington NRCSE.
Continuity and covariance Recall that for stationary processes so if C is continuous then Z is mean square continuous. To get results about sample paths.
Tch-prob1 Chapter 4. Multiple Random Variables Ex Select a student’s name from an urn. S In some random experiments, a number of different quantities.
Visual Recognition Tutorial1 Random variables, distributions, and probability density functions Discrete Random Variables Continuous Random Variables.
Nonstationary covariance structures II NRCSE. Drawbacks with deformation approach Oversmoothing of large areas Local deformations not local part of global.
Space-time processes NRCSE. Separability Separable covariance structure: Cov(Z(x,t),Z(y,s))=C S (x,y)C T (s,t) Nonseparable alternatives Temporally varying.
Using wavelet tools to estimate and assess trends in atmospheric data NRCSE.
MACHINE LEARNING 6. Multivariate Methods 1. Based on E Alpaydın 2004 Introduction to Machine Learning © The MIT Press (V1.1) 2 Motivating Example  Loan.
Space-time Modelling Using Differential Equations Alan E. Gelfand, ISDS, Duke University (with J. Duan and G. Puggioni)
Linear and generalised linear models Purpose of linear models Least-squares solution for linear models Analysis of diagnostics Exponential family and generalised.
Statistical Tools for Environmental Problems NRCSE.
(1) A probability model respecting those covariance observations: Gaussian Maximum entropy probability distribution for a given covariance observation.
Lecture II-2: Probability Review
Separate multivariate observations
EE513 Audio Signals and Systems Statistical Pattern Classification Kevin D. Donohue Electrical and Computer Engineering University of Kentucky.
Summarized by Soo-Jin Kim
Chapter 2 Dimensionality Reduction. Linear Methods
PATTERN RECOGNITION AND MACHINE LEARNING
0 Pattern Classification, Chapter 3 0 Pattern Classification All materials in these slides were taken from Pattern Classification (2nd ed) by R. O. Duda,
The horseshoe estimator for sparse signals CARLOS M. CARVALHO NICHOLAS G. POLSON JAMES G. SCOTT Biometrika (2010) Presented by Eric Wang 10/14/2010.
Time Series Analysis.
Principles of Pattern Recognition
V. Space Curves Types of curves Explicit Implicit Parametric.
Solution for non-negative ffCO2 emissions ‒ Incorporate priors ‒ Solve, using StOMP [1] ‒ StOMP solution does not give non-negative ffCO2 emissions; a.
Mathematical Preliminaries. 37 Matrix Theory Vectors nth element of vector u : u(n) Matrix mth row and nth column of A : a(m,n) column vector.
Linear Regression Andy Jacobson July 2006 Statistical Anecdotes: Do hospitals make you sick? Student’s story Etymology of “regression”
Regional climate prediction comparisons via statistical upscaling and downscaling Peter Guttorp University of Washington Norwegian Computing Center
PATTERN RECOGNITION AND MACHINE LEARNING CHAPTER 3: LINEAR MODELS FOR REGRESSION.
Basics of Neural Networks Neural Network Topologies.
1 University of Texas at Austin Machine Learning Group 图像与视频处理 计算机学院 Motion Detection and Estimation.
Week 21 Stochastic Process - Introduction Stochastic processes are processes that proceed randomly in time. Rather than consider fixed random variables.
CHAPTER 5 SIGNAL SPACE ANALYSIS
Additional Topics in Prediction Methodology. Introduction Predictive distribution for random variable Y 0 is meant to capture all the information about.
Information Geometry and Model Reduction Sorin Mitran 1 1 Department of Mathematics, University of North Carolina, Chapel Hill, NC, USA Reconstruction.
Principle Component Analysis and its use in MA clustering Lecture 12.
Statistics……revisited
Principal Component Analysis (PCA)
Adaptive Wavelet Packet Models for Texture Description and Segmentation. Karen Brady, Ian Jermyn, Josiane Zerubia Projet Ariana - INRIA/I3S/UNSA June 5,
SGPP: Spatial Gaussian Predictive Process Models for Neuroimaging Data Yimei Li Department of Biostatistics St. Jude Children’s Research Hospital Joint.
Inverse Modeling of Surface Carbon Fluxes Please read Peters et al (2007) and Explore the CarbonTracker website.
Chapter 13 Discrete Image Transforms
Space-time processes NRCSE. Separability Separable covariance structure: Cov(Z(x,t),Z(y,s))=C S (x,y)C T (s,t) Nonseparable alternatives Temporally varying.
Biointelligence Laboratory, Seoul National University
Probability Theory and Parameter Estimation I
Basic simulation methodology
Menglong Li Ph.d of Industrial Engineering Dec 1st 2016
Nonstationary covariance structures II
Spatial Point Pattern Analysis
Paul D. Sampson Peter Guttorp
Where did we stop? The Bayes decision rule guarantees an optimal classification… … But it requires the knowledge of P(ci|x) (or p(x|ci) and P(ci)) We.
EMIS 7300 SYSTEMS ANALYSIS METHODS FALL 2005
EE513 Audio Signals and Systems
Image and Video Processing
Generally Discriminant Analysis
Basis Expansions and Generalized Additive Models (2)
Multivariate Methods Berlin Chen, 2005 References:
16. Mean Square Estimation
Probabilistic Surrogate Models
Presentation transcript:

Groundwater permeability Easy to solve the forward problem: flow of groundwater given permeability of aquifer Inverse problem: determine permeability from flow (usually of tracers) With some models enough to look at first arrival of tracer at each well (breakthrough times)

Notation  is permeability b is breakthrough times expected breakthrough times Illconditioned problems: different permeabilities can yield same flow Use regularization by prior on log(  ) MRF Gaussian Convolution with MRF (discretized)

MRF prior where and n j =#{i:i~j}

Kim, Mallock & Holmes, JASA 2005 Analyzing Nonstationary Spatial Data Using Piecewise Gaussian Processes Studying oil permeability Voronoi tesselation (choose M centers from a grid) Separate power exponential in each regions

Nott & Dunsmuir, 2002, Biometrika Consider a stationary process W(s), correlation R, observed at sites s 1,..,s n. Write  (s) has covariance function

More generally Consider k independent stationary spatial fields W i (s) and a random vector Z. Write and create a nonstationary process by Its covariance (with  =Cov(Z)) is

Fig. 2. Sydney wind pattern data. Contours of equal estimated correlation with two different fixed sites, shown by open squares: (a) location 33·85°S, 151·22°E, and (b) location 33·74°S, 149·88°E. The sites marked by dots show locations of the 45 monitored sites.

Karhunen-Loéve expansion There is a unique representation of stochastic processes with uncorrelated coefficients: where the  k (s) solve and are orthogonal eigenfunctions. Example: temporal Brownian motion C(s,t)=min(s,t)  k (s)=2 1/2 sin((k-1/2)  t)/((k-1/2)  ) Conversely,

Discrete case Eigenexpansion of covariance matrix Empirically SVD of sample covariance Example: squared exponential k=1520

Tempering Stationary case: write with covariance To generalize this to a nonstationary case, use spatial powers of the k : Large  corresponds to smoother field

A simulated example

Estimating  (s) Regression spline Knots u i picked using clustering techniques Multivariate normal prior on the  ’s.

Piazza Road revisited

Tempering More fins structure More smoothness

Covariances A BCD

Karhunen-Loeve expansion revisited and where   are iid N(0, i ) Idea: use wavelet basis instead of eigenfunctions, allow for dependent  i

Spatial wavelet basis Separates out differences of averages at different scales Scaled and translated basic wavelet functions

Estimating nonstationary covariance using wavelets 2-dimensional wavelet basis obtained from two functions  and  : First generation scaled translates of all four; subsequent generations scaled translates of the detail functions. Subsequent generations on finer grids. detail functions

W-transform

Covariance expansion For covariance matrix  write Useful if D close to diagonal. Enforce by thresholding off-diagonal elements (set all zero on finest scales)

Surface ozone model ROM, daily average ozone 48 x 48 grid of 26 km x 26 km centered on Illinois and Ohio. 79 days summer x3 coarsest level (correlation length is about 300 km) Decimate leading 12 x 12 block of D by 90%, retain only diagonal elements for remaining levels.

ROM covariance

Some open questions Multivariate Kronecker structure Nonstationarity Covariates causing nonstationarity (or deterministic models) Comparison of models of nonstationarity Mean structure