We propose a novel methodology, forest floor, to visualize and interpret random forest (RF) models. RF is a popular and useful tool for non-linear multi-variate classification and regression, which yields a good trade-off between robustness (low variance) and adaptiveness (low bias). Direct interpretation of a RF model is difficult, as the explicit ensemble model of hundreds of deep trees is complex. Nonetheless, it is possible to visualize a RF model fit by its mapping from feature space to prediction space. Hereby the user is first presented with the overall geometrical shape of the model structure, and when needed one can zoom in on local details. Dimensional reduction by projection is used to visualize high dimensional shapes. The traditional method to visualize RF model structure, partial dependence plots, achieve this by averaging multiple parallel projections. We suggest to first use feature contributions, a method to decompose trees by splitting features, and then subsequently perform projections. The advantages of forest floor over partial dependence plots is that interactions are not masked by averaging. As a consequence, it is possible to locate interactions, which are not visualized in a given projection. Furthermore, we introduce: a goodness-of-visualization measure, use of colour gradients to identify interactions and an out-of-bag cross validated variant of feature contributions.

This technical note extends recent results on the computational complexity of globally minimizing the error of piecewise-affine models to the related problem of minimizing the error of switching linear regression models. In particular, we show that, on the one hand the problem is NP-hard, but on the other hand, it admits a polynomial-time algorithm with respect to the number of data points for any fixed data dimension and number of modes.

Prediction tasks over nodes and edges in networks require careful effort in engineering features used by learning algorithms. Recent research in the broader field of representation learning has led to significant progress in automating prediction by learning the features themselves. However, present feature learning approaches are not expressive enough to capture the diversity of connectivity patterns observed in networks. Here we propose node2vec, an algorithmic framework for learning continuous feature representations for nodes in networks. In node2vec, we learn a mapping of nodes to a low-dimensional space of features that maximizes the likelihood of preserving network neighborhoods of nodes. We define a flexible notion of a node's network neighborhood and design a biased random walk procedure, which efficiently explores diverse neighborhoods. Our algorithm generalizes prior work which is based on rigid notions of network neighborhoods, and we argue that the added flexibility in exploring neighborhoods is the key to learning richer representations. We demonstrate the efficacy of node2vec over existing state-of-the-art techniques on multi-label classification and link prediction in several real-world networks from diverse domains. Taken together, our work represents a new way for efficiently learning state-of-the-art task-independent representations in complex networks.

Community-based Question Answering (CQA) sites play an important role in addressing health information needs. However, a significant number of posted questions remain unanswered. Automatically answering the posted questions can provide a useful source of information for online health communities. In this study, we developed an algorithm to automatically answer health-related questions based on past questions and answers (QA). We also aimed to understand information embedded within online health content that are good features in identifying valid answers. Our proposed algorithm uses information retrieval techniques to identify candidate answers from resolved QA. In order to rank these candidates, we implemented a semi-supervised leaning algorithm that extracts the best answer to a question. We assessed this approach on a curated corpus from Yahoo! Answers and compared against a rule-based string similarity baseline. On our dataset, the semi-supervised learning algorithm has an accuracy of 86.2%. UMLS-based (health-related) features used in the model enhance the algorithm's performance by proximately 8 %. A reasonably high rate of accuracy is obtained given that the data is considerably noisy. Important features distinguishing a valid answer from an invalid answer include text length, number of stop words contained in a test question, a distance between the test question and other questions in the corpus as well as a number of overlapping health-related terms between questions. Overall, our automated QA system based on historical QA pairs is shown to be effective according to the data set in this case study. It is developed for general use in the health care domain which can also be applied to other CQA sites.

There is a widespread need for effective forecasting of financial risk using readily available financial measures, but the complicated environment facing financial practitioners and business institutions makes this very challenging. The concept of financial volatility, a required parameter for pricing many kinds of financial assets and derivatives, is critical, because it is widely expected that financial volatility implies financial risk. Therefore, accurate prediction of financial volatility is extremely important. Efficient prediction of financial volatility has been an extremely difficult task, but we can now offer a scalable and customizable mathematical model to achieve this goal, employing two approaches to forecast the volatility using financial information available online. First, we carry out a comparative study between two different machine-learning techniques -- artificial neural networks (ANN) and support vector machines (SVM) -- to forecast trading volume volatility.

So you know the Bayes rule. How does it relate to machine learning? It can be quite difficult to grasp how the puzzle pieces fit together - we know it took us a while. This article is an introduction we wish we had back then. While we have some grasp on the matter, we're not experts, so the following might contain inaccuracies or even outright errors. Feel free to point them out, either in the comments or privately.

We consider the problem of approximate joint triangularization of a set of noisy jointly diagonalizable real matrices. Approximate joint triangularizers are commonly used in the estimation of the joint eigenstructure of a set of matrices, with applications in signal processing, linear algebra, and tensor decomposition. By assuming the input matrices to be perturbations of noise-free, simultaneously diagonalizable ground-truth matrices, the approximate joint triangularizers are expected to be perturbations of the exact joint triangularizers of the ground-truth matrices. We provide a priori and a posteriori perturbation bounds on the `distance' between an approximate joint triangularizer and its exact counterpart. The a priori bounds are theoretical inequalities that involve functions of the ground-truth matrices and noise matrices, whereas the a posteriori bounds are given in terms of observable quantities that can be computed from the input matrices. From a practical perspective, the problem of finding the best approximate joint triangularizer of a set of noisy matrices amounts to solving a nonconvex optimization problem. We show that, under a condition on the noise level of the input matrices, it is possible to find a good initial triangularizer such that the solution obtained by any local descent-type algorithm has certain global guarantees. Finally, we discuss the application of approximate joint matrix triangularization to canonical tensor decomposition and we derive novel estimation error bounds.

In high dimensions we propose and analyze an aggregation estimator of the precision matrix for Gaussian graphical models. This estimator, called graphical Exponential Screening (gES), linearly combines a suitable set of individual estimators with different underlying graphs, and balances the estimation error and sparsity. We study the risk of this aggregation estimator and show that it is comparable to that of the best estimator based on a single graph, chosen by an oracle. Numerical performance of our method is investigated using both simulated and real datasets, in comparison with some state-of-art estimation procedures.

Softmax Regression (synonyms: Multinomial Logistic, Maximum Entropy Classifier, or just Multi-class Logistic Regression) is a generalization of logistic regression that we can use for multi-class classification (under the assumption that the classes are mutually exclusive). In contrast, we use the (standard) Logistic Regression model in binary classification tasks. Now, let me briefly explain how that works and how softmax regression differs from logistic regression. As the name suggests, in softmax regression (SMR), we replace the sigmoid logistic function by the so-called softmax function?: Now, this softmax function computes the probability that this training sample x(i) belongs to class j given the weight and net input z(i). So, we compute the probability p(y j x(i); wj) for each class label in j 1, ..., k.

The active course run for Statistical Learning has ended, but the course is now available in a self paced mode. You are welcome to join the course and work through the material and exercises at your own pace. When you have completed the exercises with a score of 50% or higher, you can generate your Statement of Accomplishment from within the course. The course will remain available for an extended period of time. We anticipate the content will be available until at least August 2, 2017.