1  5
Number of results to display per page
 Efron, Bradley author.
 Cambridge : Cambridge University Press, 2016.
 Description
 Book — 1 online resource (491 pages) : digital, PDF file(s).
 Summary

 Part I. Classic Statistical Inference:
 1. Algorithms and inference
 2. Frequentist inference
 3. Bayesian inference
 4. Fisherian inference and maximum likelihood estimation
 5. Parametric models and exponential families Part II. Early ComputerAge Methods:
 6. Empirical Bayes
 7. JamesStein estimation and ridge regression
 8. Generalized linear models and regression trees
 9. Survival analysis and the EM algorithm
 10. The jackknife and the bootstrap
 11. Bootstrap confidence intervals
 12. Crossvalidation and Cp estimates of prediction error
 13. Objective Bayes inference and Markov chain Monte Carlo
 14. Statistical inference and methodology in the postwar era Part III. TwentyFirst Century Topics:
 15. Largescale hypothesis testing and false discovery rates
 16. Sparse modeling and the lasso
 17. Random forests and boosting
 18. Neural networks and deep learning
 19. Supportvector machines and kernel methods
 20. Inference after model selection
 21. Empirical Bayes estimation strategies Epilogue References Index.
 (source: Nielsen Book Data)
(source: Nielsen Book Data) 9781107149892 20160823
 Efron, Bradley, author.
 New York, NY : Cambridge University Press, 2016.
 Description
 Book — xix, 475 pages : color illustrations ; 24 cm.
 Summary

 Part I. Classic Statistical Inference:
 1. Algorithms and inference
 2. Frequentist inference
 3. Bayesian inference
 4. Fisherian inference and maximum likelihood estimation
 5. Parametric models and exponential families Part II. Early ComputerAge Methods:
 6. Empirical Bayes
 7. JamesStein estimation and ridge regression
 8. Generalized linear models and regression trees
 9. Survival analysis and the EM algorithm
 10. The jackknife and the bootstrap
 11. Bootstrap confidence intervals
 12. Crossvalidation and Cp estimates of prediction error
 13. Objective Bayes inference and Markov chain Monte Carlo
 14. Statistical inference and methodology in the postwar era Part III. TwentyFirst Century Topics:
 15. Largescale hypothesis testing and false discovery rates
 16. Sparse modeling and the lasso
 17. Random forests and boosting
 18. Neural networks and deep learning
 19. Supportvector machines and kernel methods
 20. Inference after model selection
 21. Empirical Bayes estimation strategies Epilogue References Index.
 (source: Nielsen Book Data)
(source: Nielsen Book Data) 9781107149892 20170821
 Online
Online 3. Topics in statistical learning with a focus on largescale data [2018]
 Le, Ya, author.
 [Stanford, California] : [Stanford University], 2018.
 Description
 Book — 1 online resource.
 Summary

The widespread of modern information technologies to all spheres of society leads to a dramatic increase of data flow, including the formation of "big data" phenomenon. Big data are data on a massive scale in terms of volume, intensity, and complexity that exceed the ability of traditional statistical methods and standard tools. When the size of the data becomes extremely large, it may be too long to run the computing task, and even infeasible to store all of the data on a single computer. Therefore, it is necessary to turn to distributed architectures and scalable statistical methods. Big data vary in shape and call for different approaches. One type of big data is the tall data, i.e., a very large number of samples but not too many features. Chapter 1 describes a general communicationefficient algorithm for distributed statistical learning on this type of big data. Our algorithm distributes the samples uniformly to multiple machines, and uses a common reference data to improve the performance of local estimates. Our algorithm enables potentially much faster analysis, at a small cost to statistical performance. Another type of big data is the wide data, i.e., too many features but a limited number of samples. It is also called highdimensional data, to which many classical statistical methods are not applicable. Chapter 2 discusses a method of dimensionality reduction for highdimensional classification. Our method partitions features into independent communities and splits the original classification problem into separate iv smaller ones. It enables parallel computing and produces more interpretable results. For unsupervised learning methods like principle component analysis and clustering, the key challenges are choosing the optimal tuning parameter and evaluating method performance. Chapter 3 proposes a general crossvalidation approach for unsupervised learning methods. This approach randomly partitions the data matrix into K unstructured folds. For each fold, it fits a matrix completion algorithm to the rest K − 1 folds and evaluates the prediction on the holdout fold. Our approach provides a unified framework for parameter tuning in unsupervised learning, and shows strong performance in practice.
 Also online at

Online 4. Supervised evaluation of representations [electronic resource] [2017]
 Zheng, Charles.
 2017.
 Description
 Book — 1 online resource.
 Summary

Intuitively speaking, a good data representation (that is, a dimensionalityreducing mapping) reveals meaningful differences between inputs, while being relatively invariant to differences between inputs due to irrelevant factors or noise. In this work, we consider criteria for formally defining the quality of a representation, which all make use of the availability of a response variable Y to distinguish between meaningful and meaningless variation. Hence, these are criteria for supervised evaluation of representations. We consider three particular criteria: the mutual information between the representation and the response, the average accuracy of a randomized classification task, and the identification accuracy. We discuss methods for estimating all three quantities, and also show how these three quantities are interrelated. Besides the application of evaluating representations, our work also has relevance for estimation of mutual information in highdimensional data, for obtaining performance guarantees for recognition systems, and for making statements about the generalizability of certain kinds of classificationbased experiments which are found in neuroscience.
 Also online at

Special Collections
Special Collections  Status 

University Archives  Request onsite access 
3781 2017 Z  Inlibrary use 
Online 5. Causal inference with random forests [electronic resource] [2016]
 Wager, Stefan.
 2016.
 Description
 Book — 1 online resource.
 Summary

Random forests, introduced by Breiman [2001], have become one of the most popular machine learning algorithms among practitioners, and reliably achieve good predictive performance across several application areas. This has led to considerable interest in using random forests for doing science, or drawing statistical inferences in problems that do not reduce immediately to prediction. As a step in this direction, this thesis studies how random forests can be used for understanding treatment effect heterogeneity as it may arise in, e.g., personalized medicine. Our main contributions are as follows:  We develop a causal forest algorithm for heterogeneous treatment effect estimation, and find our method to be substantially more powerful at identifying treatment heterogeneity than traditional methods based on nearestneighbor matching, especially when the number of considered covariates is large.  We provide an asymptotic statistical analysis of causal forests, and prove a Gaussian limit result. We then propose a practical method for estimating the noise scale of causal forests, thus allowing for valid statistical inference with causal forests.  In a highdimensional regime where the problem complexity and the number of observations jointly approach infinity, we identify the signal strength at which treebased methods become able to accurately detect treatment heterogeneity. Perhaps strikingly, we find that the required signal strength only scales logarithmically in the dimension of the problem. Taken together, these results show that random forests  despite often being understood as a mere black box predictive algorithm  provide a powerful toolbox for heterogeneous treatment effect estimation in modern largescale problems.
 Also online at

Special Collections
Special Collections  Status 

University Archives  Request onsite access 
3781 2016 W  Inlibrary use 
Articles+
Journal articles, ebooks, & other eresources
 Articles+ results include