Prior design is one of the most important problems in both statistics and machine learning. The cross validation (CV) and the widely applicable information criterion (WAIC) are predictive measures of the Bayesian estimation, however, it has been difficult to apply them to find the optimal prior because their mathematical properties in prior evaluation have been unknown and the region of the hyperparameters is too wide to be examined. In this paper, we derive a new formula by which the theoretical relation among CV, WAIC, and the generalization loss is clarified and the optimal hyperparameter can be directly found. By the formula, three facts are clarified about predictive prior design. Firstly, CV and WAIC have the same second order asymptotic expansion, hence they are asymptotically equivalent to each other as the optimizer of the hyperparameter. Secondly, the hyperparameter which minimizes CV or WAIC makes the average generalization loss to be minimized asymptotically but does not the random generalization loss. And lastly, by using the mathematical relation between priors, the variances of the optimized hyperparameters by CV and WAIC are made smaller with small computational costs. Also we show that the optimized hyperparameter by DIC or the marginal likelihood does not minimize the average or random generalization loss in general.

We focus on an interpolation method referred to Bayesian reconstruction in this paper. Whereas in standard interpolation methods missing data are interpolated deterministically, in Bayesian reconstruction, missing data are interpolated probabilistically using a Bayesian treatment. In this paper, we address the framework of Bayesian reconstruction and its application to the traffic data reconstruction problem in the field of traffic engineering. In the latter part of this paper, we describe the evaluation of the statistical performance of our Bayesian traffic reconstruction model using a statistical mechanical approach and clarify its statistical behavior.

Neuroimage analysis usually involves learning thousands or even millions of variables using only a limited number of samples. In this regard, sparse models, e.g. the lasso, are applied to select the optimal features and achieve high diagnosis accuracy. The lasso, however, usually results in independent unstable features. Stability, a manifest of reproducibility of statistical results subject to reasonable perturbations to data and the model, is an important focus in statistics, especially in the analysis of high dimensional data. In this paper, we explore a nonnegative generalized fused lasso model for stable feature selection in the diagnosis of Alzheimer's disease. In addition to sparsity, our model incorporates two important pathological priors: the spatial cohesion of lesion voxels and the positive correlation between the features and the disease labels. To optimize the model, we propose an efficient algorithm by proving a novel link between total variation and fast network flow algorithms via conic duality. Experiments show that the proposed nonnegative model performs much better in exploring the intrinsic structure of data via selecting stable features compared with other state-of-the-arts.

We study adaptive data-dependent dimensionality reduction in the context of supervised learning in general metric spaces. Our main statistical contribution is a generalization bound for Lipschitz functions in metric spaces that are doubling, or nearly doubling. On the algorithmic front, we describe an analogue of PCA for metric spaces: namely an efficient procedure that approximates the data's intrinsic dimension, which is often much lower than the ambient dimension. Our approach thus leverages the dual benefits of low dimensionality: (1) more efficient algorithms, e.g., for proximity search, and (2) more optimistic generalization bounds.

Hierarchical probabilistic models, such as Gaussian mixture models, are widely used for unsupervised learning tasks. These models consist of observable and latent variables, which represent the observable data and the underlying data-generation process, respectively. Unsupervised learning tasks, such as cluster analysis, are regarded as estimations of latent variables based on the observable ones. The estimation of latent variables in semi-supervised learning, where some labels are observed, will be more precise than that in unsupervised, and one of the concerns is to clarify the effect of the labeled data. However, there has not been sufficient theoretical analysis of the accuracy of the estimation of latent variables. In a previous study, a distribution-based error function was formulated, and its asymptotic form was calculated for unsupervised learning with generative models. It has been shown that, for the estimation of latent variables, the Bayes method is more accurate than the maximum-likelihood method. The present paper reveals the asymptotic forms of the error function in Bayesian semi-supervised learning for both discriminative and generative models. The results show that the generative model, which uses all of the given data, performs better when the model is well specified.

While catchphrases such as big data, smart data, data-intensive science, or smart dust highlight different aspects, they share a common theme: Namely, a shift towards a data-centric perspective in which the synthesis and analysis of data at an ever-increasing spatial, temporal, and thematic resolution promises new insights, while, at the same time, reducing the need for strong domain theories as starting points. In terms of the envisioned methodologies, those catchphrases tend to emphasize the role of predictive analytics, that is, statistical techniques including data mining and machine learning, as well as supercomputing. Interestingly, however, while this perspective takes the availability of data as a given, it does not answer the question how one would discover the required data in today’s chaotic information universe, how one would understand which datasets can be meaningfully integrated, and how to communicate the results to humans and machines alike. The semantic web addresses these questions. In the following, we argue why the data train needs semantic rails. We point out that making sense of data and gaining new insights works best if inductive and deductive techniques go hand-in-hand instead of competing over the prerogative of interpretation.

Many information extraction and knowledge base construction systems are addressing the challenge of deriving knowledge from text. A key problem in constructing these knowledge bases from sources like the web is overcoming the erroneous and incomplete information found in millions of candidate extractions. To solve this problem, we turn to semantics — using ontological constraints between candidate facts to eliminate errors. In this article, we represent the desired knowledge base as a knowledge graph and introduce the problem of knowledge graph identification, collectively resolving the entities, labels, and relations present in the knowledge graph. Knowledge graph identification requires reasoning jointly over millions of extractions simultaneously, posing a scalability challenge to many approaches. We use probabilistic soft logic (PSL), a recently-introduced statistical relational learning framework, to implement an efficient solution to knowledge graph identification and present state-of-the-art results for knowledge graph construction while performing an order of magnitude faster than competing methods.

Deep learning can achieve outstanding results in various fields. However, it requires so significant computational power that graphics processing units (GPUs) and/or numerous computers are often required for the practical application. We have developed a new distributed calculation framework called "Sashimi" that allows any computer to be used as a distribution node only by accessing a website. We have also developed a new JavaScript neural network framework called "Sukiyaki" that uses general purpose GPUs with web browsers. Sukiyaki performs 30 times faster than a conventional JavaScript library for deep convolutional neural networks (deep CNNs) learning. The combination of Sashimi and Sukiyaki, as well as new distribution algorithms, demonstrates the distributed deep learning of deep CNNs only with web browsers on various devices. The libraries that comprise the proposed methods are available under MIT license at http://mil-tokyo.github.io/.

Scientists in many fields have the common and basic need of dimensionality reduction: visualizing the underlying structure of the massive multivariate data in a low-dimensional space. However, many dimensionality reduction methods confront the so-called "crowding problem" that clusters tend to overlap with each other in the embedding. Previously, researchers expect to avoid that problem and seek to make clusters maximally separated in the embedding. However, the proposed in-tree (IT) based method, called IT-map, allows clusters in the embedding to be locally overlapped, while seeking to make them distinguishable by some small yet key parts. IT-map provides a simple, effective and novel solution to cluster-preserving mapping, which makes it possible to cluster the original data points interactively and thus should be of general meaning in science and engineering.

In our physically inspired in-tree (IT) based clustering algorithm and the series after it, there is only one free parameter involved in computing the potential value of each point. In this work, based on the Delaunay Triangulation or its dual Voronoi tessellation, we propose a nonparametric process to compute potential values by the local information. This computation, though nonparametric, is relatively very rough, and consequently, many local extreme points will be generated. However, unlike those gradient-based methods, our IT-based methods are generally insensitive to those local extremes. This positively demonstrates the superiority of these parametric (previous) and nonparametric (in this work) IT-based methods.