We present a probabilistic framework for learning with heterogeneous multiview data where some views are given as ordinal, binary, or real-valued feature matrices, and some views as similarity matrices. Our framework has the following distinguishing aspects: (i) a unified latent factor model for integrating information from diverse feature (ordinal, binary, real) and similarity based views, and predicting the missing data in each view, leveraging view correlations; (ii) seamless adaptation to binary/multiclass classification where data consists of multiple feature and/or similarity-based views; and (iii) an efficient, variational inference algorithm which is especially flexible in modeling the views with ordinal-valued data (by learning the cutpoints for the ordinal data), and extends naturally to streaming data settings. Our framework subsumes methods such as multiview learning and multiple kernel learning as special cases. We demonstrate the effectiveness of our framework on several real-world and benchmarks datasets.

Lasso is a popular method for high-dimensional variable selection, but it hinges on a tuning parameter that is difficult to calibrate in practice. In this study, we introduce TREX, an alternative to Lasso with an inherent calibration to all aspects of the model. This adaptation to the entire model renders TREX an estimator that does not require any calibration of tuning parameters. We show that TREX can outperform cross-validated Lasso in terms of variable selection and computational efficiency. We also introduce a bootstrapped version of TREX that can further improve variable selection. We illustrate the promising performance of TREX both on synthetic data and on two biological data sets from the fields of genomics and proteomics.

Curriculum learning (CL) or self-paced learning (SPL) represents a recently proposed learning regime inspired by the learning process of humans and animals that gradually proceeds from easy to more complex samples in training. The two methods share a similar conceptual learning paradigm, but differ in specific learning schemes. In CL, the curriculum is predetermined by prior knowledge, and remain fixed thereafter. Therefore, this type of method heavily relies on the quality of prior knowledge while ignoring feedback about the learner. In SPL, the curriculum is dynamically determined to adjust to the learning pace of the leaner. However, SPL is unable to deal with prior knowledge, rendering it prone to overfitting. In this paper, we discover the missing link between CL and SPL, and propose a unified framework named self-paced curriculum leaning (SPCL). SPCL is formulated as a concise optimization problem that takes into account both prior knowledge known before training and the learning progress during training. In comparison to human education, SPCL is analogous to "instructor-student-collaborative" learning mode, as opposed to "instructor-driven" in CL or "student-driven" in SPL. Empirically, we show that the advantage of SPCL on two tasks.

We develop the Dynamic Chinese Restaurant Process (DCRP) which incorporates time-evolutionary feature in dependent Dirichlet Process mixture models. This model can capture the dynamic change of mixture components, allowing clusters to emerge, vanish and vary over time. All these macroscopic changes are controlled by tracing the birth and death of every single element. We investigate the  properties of dependent Dirichlet Process mixture model based on DCRP and develop corresponding Gibbs Sampler for posterior inference. We also conduct simulation and empirical studies to compare this model with traditional CRP and related models. The results show that this model can provide better results for sequential data, especially for data with heterogeneous lifetime distribution.

Online learning from imbalanced streaming data to capture the nonlinearity and heterogeneity of the data is significant in machine learning and data mining. To tackle this problem, we propose a kernelized online imbalanced learning (KOIL) algorithm to directly maximize the area under the ROC curve (AUC). We address two more challenges: 1) How to control the number of support vectors without sacrificing model performance; and 2) how to restrict the fluctuation of the learned decision function to attain smooth updating. To this end, we introduce two buffers with fixed budgets (buffer sizes) for positive class and negative class, respectively, to store the learned support vectors, which can allow us to capture the global information of the decision boundary. When determining the weight of a new support vector, we confine its influence only to its $k$-nearest opposite support vectors. This can restrict the effect of new instances and prevent the harm of outliers. More importantly, we design a sophisticated scheme to compensate the model after replacement is conducted when either buffer is full. With this compensation, the learned model approaches the one learned with infinite budgets. We present both theoretical analysis and extensive experimental comparison to demonstrate the effectiveness of our proposed KOIL.

In multi-task learning (MTL), multiple related tasks are learned jointly by sharing information across them. Many MTL algorithms have been proposed to learn the underlying task groups. However, those methods are limited to learn the task groups at only a single level, which may be not sufficient to model the complex structure among tasks in many real-world applications. In this paper, we propose a Multi-Level Task Grouping (MeTaG) method to learn the multi-level grouping structure instead of only one level among tasks. Specifically, by assuming the number of levels to be H, we decompose the parameter matrix into a sum of H component matrices, each of which is regularized with a l2 norm on the pairwise difference among parameters of all the tasks to construct level-specific task groups. For optimization, we employ the smoothing proximal gradient method to efficiently solve the objective function of the MeTaG model. Moreover, we provide theoretical analysis to show that under certain conditions the MeTaG model can recover the true parameter matrix and the true task groups in each level with high probability. We experiment our approach on both synthetic and real-world datasets, showing competitive performance over state-of-the-art MTL methods.

Feature grouping has been demonstrated to be promising in learning with high-dimensional data. It helps reduce the variances in the estimation and improves the stability of feature selection. One major limitation of existing feature grouping approaches is that some similar but different feature groups are often mis-fused, leading to impaired performance. In this paper, we propose a Discriminative Feature Grouping (DFG) method to discover the feature groups with enhanced discrimination. Different from existing methods, DFG adopts a novel regularizer for the feature coefficients to trade-off between fusing and discriminating feature groups. The proposed regularizer consists of a ell_1 norm to enforce feature sparsity and a pairwise ell_infty norm to encourage the absolute differences among any three feature coefficients to be similar. To achieve better asymptotic property, we generalize the proposed regularizer to an adaptive one where the feature coefficients are weighted based on the solution of some estimator with root-n consistency. For optimization, we employ the alternating direction method of multipliers to solve the proposed methods efficiently. Experimental results on synthetic and real-world datasets demonstrate that the proposed methods have good performance compared with the state-of-the-art feature grouping methods.

Graphical models provide a rich framework for summarizing the dependencies among variables. The graphical lasso approach attempts to learn the structure of a Gaussian graphical model (GGM) by maximizing the log likelihood of the data, subject to an l1 penalty on the elements of the inverse covariance matrix. Most algorithms for solving the graphical lasso problem do not scale to a very large number of variables. Furthermore, the learned network structure is hard to interpret. To overcome these challenges, we propose a novel GGM structure learning method that exploits the fact that for many real-world problems we have prior knowledge that certain edges are unlikely to be present. For example, in gene regulatory networks, a pair of genes that does not participate together in any of the cellular processes, typically referred to as pathways, is less likely to be connected. In computer vision applications in which each variable corresponds to a pixel, each variable is likely to be connected to the nearby variables. In this paper, we propose the pathway graphical lasso, which learns the structure of a GGM subject to pathway-based constraints. In order to solve this problem, we decompose the network into smaller parts, and use a message-passing algorithm in order to communicate among the subnetworks. Our algorithm has orders of magnitude improvement in run time compared to the state-of-the-art optimization methods for the graphical lasso problem that were modified to handle pathway-based constraints.

Topic modeling is a powerful tool for uncovering latent structure in many domains, including medicine, finance, and vision. The goals for the model vary depending on the application: sometimes the discovered topics are used for prediction or another downstream task. In other cases, the content of the topic may be of intrinsic scientific interest. Unfortunately, even when one uses modern sparse techniques, discovered topics are often difficult to interpret due to the high dimensionality of the underlying space. To improve topic interpretability, we introduce Graph-Sparse LDA, a hierarchical topic model that uses knowledge of relationships between words (e.g., as encoded by an ontology). In our model, topics are summarized by a few latent concept-words from the underlying graph that explain the observed words. Graph-Sparse LDA recovers sparse, interpretable summaries on two real-world biomedical datasets while matching state-of-the-art prediction performance.

Much of the focus on finding good representations in reinforcement learning has been on learning complex non-linear predictors of value. Policy gradient algorithms, which directly represent the policy, often need fewer parameters to learn good policies. However, they typically employ a fixed parametric representation that may not be sufficient for complex domains. This paper introduces the Policy Tree algorithm, which can learn an adaptive representation of policy in the form of a decision tree over different instantiations of a base policy. Policy gradient is used both to optimize the parameters and to grow the tree by choosing splits that enable the maximum local increase in the expected return of the policy. Experiments show that this algorithm can choose genuinely helpful splits and significantly improve upon the commonly used linear Gibbs softmax policy, which we choose as our base policy.