We examine the effect of the Group Lasso (gLasso) regularizer in selecting the salient nodes of Deep Neural Network (DNN) hidden layers by applying a DNN-HMM hybrid speech recognizer to TED Talks speech data. We test two types of gLasso regularization, one for outgoing weight vectors and another for incoming weight vectors, as well as two sizes of DNNs: 2048 hidden layer nodes and 4096 nodes. Furthermore, we compare gLasso and L2 regularizers. Our experiment results demonstrate that our DNN training, in which the gLasso regularizer was embedded, successfully selected the hidden layer nodes that are necessary and sufficient for achieving high classification power.

Embeddings are ubiquitous in machine learning, appearing in recommender systems, NLP, and many other applications. Researchers and developers often need to explore the properties of a specific embedding, and one way to analyze embeddings is to visualize them. We present the Embedding Projector, a tool for interactive visualization and interpretation of embeddings.

In this paper we combine two important extensions of ordinary least squares regression: regularization and optimal scaling. Optimal scaling (sometimes also called optimal scoring) has originally been developed for categorical data, and the process finds quantifications for the categories that are optimal for the regression model in the sense that they maximize the multiple correlation. Although the optimal scaling method was developed initially for variables with a limited number of categories, optimal transformations of continuous variables are a special case. We will consider a variety of transformation types; typically we use step functions for categorical variables, and smooth (spline) functions for continuous variables. Both types of functions can be restricted to be monotonic, preserving the ordinal information in the data. In addition to optimal scaling, three regularization methods will be considered: Ridge regression, the Lasso, and the Elastic Net. The resulting method will be called ROS Regression (Regularized Optimal Scaling Regression. We will show that the basic OS algorithm provides straightforward and efficient estimation of the regularized regression coefficients, automatically gives the Group Lasso and Blockwise Sparse Regression, and extends them with monotonicity properties. We will show that Optimal Scaling linearizes nonlinear relationships between predictors and outcome, and improves upon the condition of the predictor correlation matrix, increasing (on average) the conditional independence of the predictors. Alternative options for regularization of either regression coefficients or category quantifications are mentioned. Extended examples are provided. Keywords: Categorical Data, Optimal Scaling, Conditional Independence, Step Functions, Splines, Monotonic Transformations, Regularization, Lasso, Elastic Net, Group Lasso, Blockwise Sparse Regression.

With the large volume of new information created every day, determining the validity of information in a knowledge graph and filling in its missing parts are crucial tasks for many researchers and practitioners. To address this challenge, a number of knowledge graph completion methods have been developed using low-dimensional graph embeddings. Although researchers continue to improve these models using an increasingly complex feature space, we show that simple changes in the architecture of the underlying model can outperform state-of-the-art models without the need for complex feature engineering. In this work, we present a shared variable neural network model called ProjE that fills-in missing information in a knowledge graph by learning joint embeddings of the knowledge graph's entities and edges, and through subtle, but important, changes to the standard loss function. In doing so, ProjE has a parameter size that is smaller than 11 out of 15 existing methods while performing 37% better than the current-best method on standard datasets. We also show, via a new fact checking task, that ProjE is capable of accurately determining the veracity of many declarative statements. Knowledge Graphs (KGs) have become a crucial resource for many tasks in machine learning, data mining, and artificial intelligence applications including question answering [34], entity disambiguation [7], named entity linking [14], fact checking [32], and link prediction [28] to name a few.

Semidefinite programs have recently been developed for the problem of community detection, which may be viewed as a special case of the stochastic blockmodel. Here, we develop a semidefinite program that can be tailored to other instances of the blockmodel, such as non-assortative networks and overlapping communities. We establish label recovery in sparse settings, with conditions that are analogous to recent results for community detection. In settings where the data is not generated by a blockmodel, we give an oracle inequality that bounds excess risk relative to the best blockmodel approximation. Simulations are presented for community detection, for overlapping communities, and for latent space models.

Previous research has shown that computation of convolution in the frequency domain provides a significant speedup versus traditional convolution network implementations. However, this performance increase comes at the expense of repeatedly computing the transform and its inverse in order to apply other network operations such as activation, pooling, and dropout. We show, mathematically, how convolution and activation can both be implemented in the frequency domain using either the Fourier or Laplace transformation. The main contributions are a description of spectral activation under the Fourier transform and a further description of an efficient algorithm for computing both convolution and activation under the Laplace transform. By computing both the convolution and activation functions in the frequency domain, we can reduce the number of transforms required, as well as reducing overall complexity. Our description of a spectral activation function, together with existing spectral analogs of other network functions may then be used to compose a fully spectral implementation of a convolution network.

Any observer can sense the artistic style of painting, even if it takes training to articulate it. To an art historian, the artistic style is the primary means of classifying the painting [10]. However, artistic style is not well defined, and may be loosely described as ".. a distinctive manner which permits the grouping of works into related categories" [1]. Algorithmically determining the artistic style of an artwork is a challenging problem which may include analysis of features such as the painting's color, its texture, and its subject matter, or none of those at all. Detecting the style of a digitized image of a painting poses additional challenges raised by the digitization process, which itself has consequences that may affect the ability of a machine to correctly detect artistic style; for instance, textures may be affected by the resolution of the digitization. Despite these challenges, intelligent systems for detecting artistic style would be useful for identification and retrieval of images of a similar style. In this paper we investigate several methods based on recent advances in convolutional neural networks for large-scale determination of artistic style. In particular, we adapt the neural-style algorithm introduced in [2] for large-scale style classification, showing performance that is competitive with other deep convolutional neural network based approaches. 1 Figure 1: Original image on the left, after application of the'neural-style' algorithm (style image'Starry Night', by Van Gogh) on the right.

V ariational autoencoders (VAEs), that are built upon deep neural networks have emerged as popular generative models in computer vision. Most of the work towards improving variational autoencoders has focused mainly on making the approximations to the posterior flexible and accurate, leading to tremendous progress. However, there have been limited efforts to replace pixel-wise reconstruction, which have known shortcomings. In this work, we use real-valued non-volume preserving transformations (real NVP) to exactly compute the conditional likelihood of the data given the latent distribution. W e show that a simple VAE with this form of reconstruction is competitive with complicated VAE structures, on image modeling tasks. As part of our model, we develop powerful conditional coupling layers that enable real NVP to learn with fewer intermediate layers.

Critically ill patients in regular wards are vulnerable to unanticipated clinical dete- rioration which requires timely transfer to the intensive care unit (ICU). To allow for risk scoring and patient monitoring in such a setting, we develop a novel Semi- Markov Switching Linear Gaussian Model (SSLGM) for the inpatients' physiol- ogy. The model captures the patients' latent clinical states and their corresponding observable lab tests and vital signs. We present an efficient unsupervised learn- ing algorithm that capitalizes on the informatively censored data in the electronic health records (EHR) to learn the parameters of the SSLGM; the learned model is then used to assess the new inpatients' risk for clinical deterioration in an online fashion, allowing for timely ICU admission. Experiments conducted on a het- erogeneous cohort of 6,094 patients admitted to a large academic medical center show that the proposed model significantly outperforms the currently deployed risk scores such as Rothman index, MEWS, SOFA and APACHE.

We consider the following general hidden hubs model: an $n \times n$ random matrix $A$ with a subset $S$ of $k$ special rows (hubs): entries in rows outside $S$ are generated from the probability distribution $p_0 \sim N(0,\sigma_0^2)$; for each row in $S$, some $k$ of its entries are generated from $p_1 \sim N(0,\sigma_1^2)$, $\sigma_1>\sigma_0$, and the rest of the entries from $p_0$. The problem is to identify the high-degree hubs efficiently. This model includes and significantly generalizes the planted Gaussian Submatrix Model, where the special entries are all in a $k \times k$ submatrix. There are two well-known barriers: if $k\geq c\sqrt{n\ln n}$, just the row sums are sufficient to find $S$ in the general model. For the submatrix problem, this can be improved by a $\sqrt{\ln n}$ factor to $k \ge c\sqrt{n}$ by spectral methods or combinatorial methods. In the variant with $p_0=\pm 1$ (with probability $1/2$ each) and $p_1\equiv 1$, neither barrier has been broken. We give a polynomial-time algorithm to identify all the hidden hubs with high probability for $k \ge n^{0.5-\delta}$ for some $\delta >0$, when $\sigma_1^2>2\sigma_0^2$. The algorithm extends to the setting where planted entries might have different variances each at least as large as $\sigma_1^2$. We also show a nearly matching lower bound: for $\sigma_1^2 \le 2\sigma_0^2$, there is no polynomial-time Statistical Query algorithm for distinguishing between a matrix whose entries are all from $N(0,\sigma_0^2)$ and a matrix with $k=n^{0.5-\delta}$ hidden hubs for any $\delta >0$. The lower bound as well as the algorithm are related to whether the chi-squared distance of the two distributions diverges. At the critical value $\sigma_1^2=2\sigma_0^2$, we show that the general hidden hubs problem can be solved for $k\geq c\sqrt n(\ln n)^{1/4}$, improving on the naive row sum-based method.