Session title: New Statistical Machine Learning Tools
Organizer: Liu, Yufeng (UNC)
Chair: Liu, Yufeng (UNC)
Time: June 5th, 1:15pm – 2:45pm
Location: VEC 201
Speech 1: Inference, Computation, and Visualization for Convex Clustering and Biclustering
Speaker: Genevera Allen (Rice)
Abstract: Hierarchical clustering enjoys wide popularity because of its fast computation, ease of interpretation, and appealing visualizations via the dendogram and cluster heatmap. Recently, several have proposed and studied convex clustering and biclustering which, similar in spirit to hierarchical clustering, achieve cluster merges via convex fusion penalties. While these techniques enjoy superior statistical performance, they suffer from slower computation and are not generally conducive to representation as a dendogram. In the first part of the talk, we present new convex (bi)clustering methods and fast algorithms that inherit all of the advantages of hierarchical clustering. Specifically, we develop a new fast approximation and variation of the convex (bi)clustering solution path that can be represented as a dendogram or cluster heatmap. Also, as one tuning parameter indexes the sequence of convex (bi)clustering solutions, we can use these to develop interactive and dynamic visualization strategies that allow one to watch data form groups as the tuning parameter varies. In the second part of this talk, we consider how to conduct inference for convex clustering solutions that addresses questions like: Are there clusters in my data set? Or, should two clustersbe merged into one? To achieve this, we develop a new geometric representation of Hotelling’s T^2-test that allows us to use the selective inference paradigm to test multivariate hypotheses for the first time. We can use this approach to test hypotheses and calculate confidence ellipsoids on the cluster means resulting from convex clustering. We apply these techniques to examples from text mining and cancer genomics. This is joint work with John Nagorski, Michael Weylandt, and Frederick Campbell.
Speech 2: High-dimensional Cost-constrained Regression via Non-convex Optimization
Speaker: Guan Yu (SUNY Buffalo)
Abstract: In modern predictive modeling process, budget constraints become a very important consideration due to the high cost of collecting data using new techniques such as brain imaging and DNA sequencing. This motivates us to develop new and efficient high-dimensional costconstrained predictive modeling methods. In this paper, to address this challenge, we first study a new non-convex high-dimensional cost-constrained linear regression problem, that is, we aim to find the cost-constrained regression model with the smallest expected prediction error among all models satisfying a budget constraint. The non-convex budget constraint makes this problem NP-hard. In order to estimate the regression coefficient vector of the cost-constrained regression model, we propose a new discrete extension of recent first-order continuous optimization methods. In particular, our method delivers a series of estimates of the regression coefficient vector by solving a sequence of 0-1 knapsack problems that can be addressed by many existing algorithms such as dynamic programming efficiently. Next, we show some extensions of our proposed method for statistical learning problems using loss functions with Lipschitz continuous gradient. It can be also extended to problems with groups of variables or multiple constraints. Theoretically, we prove that the series of the estimates generated by our iterative algorithm converge to a first-order stationary point, which can be a globally optimal solution to the nonconvex high-dimensional cost-constrained regression problem. Computationally, our numerical studies show that the proposed method can solve problems of fairly high dimensions and has promising estimation, prediction, and model selection performance.
Speech 3: Modeling Hybrid Traits for Comorbidity and Genetic Studies of Alcohol and Nicotine Co-Dependence
Speaker: Heping Zhang (Yale)
Abstract: I will present a novel multivariate model for analyzing hybrid traits and identifying genetic factors for comorbid conditions. Comorbidity is a common phenomenon in mental health in which an individual suffers from multiple disorders simultaneously. For example, in the Study of Addiction: Genetics and Environment (SAGE), alcohol and nicotine addiction were recorded through multiple assessments that we refer to as hybrid traits. Statistical inference for studying the genetic basis of hybrid traits has not been well-developed. Recent rank-based methods have been utilized for conducting association analyses of hybrid traits but do not inform the strength or direction of effects. To overcome this limitation, a parametric modeling framework is imperative. Although such parametric frameworks have been proposed in theory, they are neither well-developed nor extensively used in practice due to their reliance on complicated likelihood functions that have high computational complexity. Many existing parametric frameworks tend to instead use pseudo-likelihoods to reduce computational burdens. Here, we develop a model fitting algorithm for the full likelihood. Our extensive simulation studies demonstrate that inference based on the full likelihood can control the type-I error rate, and gains power and improves the effect size estimation when compared with several existing methods for hybrid models. These advantages remain even if the distribution of the latent variables is misspecified. After analyzing the SAGE data, we identify three genetic variants (rs7672861, rs958331, rs879330) that are significantly associated with the comorbidity of alcohol and nicotine addiction at the chromosome-wide level. Moreover, our approach has greater power in this analysis than several existing methods for hybrid traits. Although the analysis of the SAGE data motivated us to develop the model, it can be broadly applied to analyze any hybrid responses.