We address the problem of detecting changes in multivariate datastreams, and we investigate the intrinsic difficulty that change-detection methods have to face when the data dimension scales. In particular, we consider a general approach where changes are detected by comparing the distribution of the log-likelihood of the datastream over different time windows. Despite the fact that this approach constitutes the frame of several change-detection methods, its effectiveness when data dimension scales has never been investigated, which is indeed the goal of our paper. We show that the magnitude of the change can be naturally measured by the symmetric Kullback-Leibler divergence between the pre- and post-change distributions, and that the detectability of a change of a given magnitude worsens when the data dimension increases. This problem, which we refer to as \emph{detectability loss}, is due to the linear relationship between the variance of the log-likelihood and the data dimension. We analytically derive the detectability loss on Gaussian-distributed datastreams, and empirically demonstrate that this problem holds also on real-world datasets and that can be harmful even at low data-dimensions (say, 10).

More than six billion dollars: That's how much health care providers and consumers will be spending every year on artificial intelligence tools by 2021--a tenfold increase from today--according to a new report from research firm Frost & Sullivan. AI will be everywhere--from diagnosing cancer to providing weight-loss coaching, says Venkat Rajan, who has the great title of global director for the company's Visionary Healthcare Program. "Prior to 2015, most of what was happening was sort of academic: pilot programs, exploratory, proof of concept-type stuff," he says. AI's ability to sort through scads of information, and remember everything it has ever seen, could enable a digital (and congenial) version of Dr. House, the brilliant diagnostician from the eponymous TV show, says Rajan. "At first, it's a complete mystery, it could be one of ten different things," he says, about the process in the show, and real life, called differential diagnosis. "And then he's able to sort through various issues, you know, illuminate certain factors on why it's not one of these other conditions, and he's able to pull something from memory that figures out ultimately what it is, and they can provide the appropriate treatment." Robots won't steal doctors' jobs, says Rajan, but they will spare overworked docs some of the dangerous fatigue that can lead to mistakes.

Due to the nonlinear but highly interpretable representations,decision tree (DT) models have significantly attracted a lot of attention of researchers. However, DT models usually suffer from the curse of dimensionality and achieve degenerated performance when there are many noisy features. To address these issues, this paper first presents a novel data-dependent generalization error bound for the perceptron decision tree(PDT), which provides the theoretical justification to learn a sparse linear hyperplane in each decision node and to prune the tree. Following our analysis, we introduce the notion of sparse perceptron decision node (SPDN) with a budget constraint on the weight coefficients, and propose a sparse perceptron decision tree (SPDT) algorithm to achieve nonlinear prediction performance. To avoid generating an unstable and complicated decision tree and improve the generalization of the SPDT, we present a pruning strategy by learning classifiers to minimize cross-validation errors on each SPDN. Extensive empirical studies verify that our SPDT is more resilient to noisy features and effectively generates a small,yet accurate decision tree. Compared with state-of-the-art DT methods and SVM, our SPDT achieves better generalization performance on ultrahigh dimensional problems with more than 1 million features.

One of the key uses of causes is to explain why things happen. Explanations of specific events, like an individual's heart attack on Monday afternoon or a particular car accident, help assign responsibility and inform our future decisions. Computational methods for causal inference make use of the vast amounts of data collected by individuals to better understand their behavior and improve their health. However, most methods for explanation of specific events have provided theoretical approaches with limited applicability. In contrast we make two main contributions: an algorithm for explanation that calculates the strength of token causes, and an evaluation based on simulated data that enables objective comparison against prior methods and ground truth. We show that the approach finds the correct relationships in classic test cases (causal chains, common cause, and backup causation) and in a realistic scenario (explaining hyperglycemic episodes in a simulation of type 1 diabetes).

Timed Failure Propagation Graphs (TFPGs) are used in the design of safety-critical systems as a way of modeling failure propagation, and to evaluate and implement diagnostic systems. TFPGs are a very rich formalism: they allow to model Boolean combinations of faults and events, also dependent on the operational modes of the system and quantitative delays between them. TFPGs are often produced manually, from a given dynamic system of greater complexity, as abstract representations of the system behavior under specific faulty conditions. In this paper we tackle two key difficulties in this process: first, how to make sure that no important behavior of the system is overlooked in the TFPG, and that no spurious, non-existent behavior is introduced; second, how to devise the correct values for the delays between events. We propose a model checking approach to automatically validate the completeness and tightness of a TFPG for a given infinite-state dynamic system, and a procedure for the automated synthesis of the delay parameters. The proposed approach is evaluated on a number of synthetic and industrial benchmarks.

Recent work has raised the challenge of efficient automated troubleshooting in domains where repairing a set of components in a single repair action is cheaper than repairing each of them separately. This corresponds to cases where there is a non-negligible overhead to initiating a repair action and to testing the system after a repair action. In this work we propose several algorithms for choosing which batch of components to repair, so as to minimize the overall repair costs. Experimentally, we show the benefit of these algorithms over repairing components one at a time.

Last episode, we treated our Decision Tree as a blackbox. In this episode, we'll build one on a real dataset, add code to visualize it, and practice reading it - so you can see how it works under the hood. And hey -- I may have gone a little fast through some parts. Just let me know, I'll slow down. Also: we'll do a Q&A episode down the road, so if anything is unclear, just ask! Subscribe to the Google Developers: http://goo.gl/mQyv5L

Model-Based Diagnosis (MBD) is a principled and domain-independent way of analyzing why a system under examination is not behaving as expected. Given an abstract description (model) of the system's components and their behavior when functioning normally, MBD techniques rely on observations about the actual system behavior to reason about possible causes when there are discrepancies between the expected and observed behavior. Due to its generality, MBD has been successfully applied in a variety of application domains over the last decades. In many application domains of MBD, testing different hypotheses about the reasons for a failure can be computationally costly, e.g., because complex simulations of the system behavior have to be performed. In this work, we therefore propose different schemes of parallelizing the diagnostic reasoning process in order to better exploit the capabilities of modern multi-core computers. We propose and systematically evaluate parallelization schemes for Reiter's hitting set algorithm for finding all or a few leading minimal diagnoses using two different conflict detection techniques. Furthermore, we perform initial experiments for a basic depth-first search strategy to assess the potential of parallelization when searching for one single diagnosis. Finally, we test the effects of parallelizing "direct encodings" of the diagnosis problem in a constraint solver.

A new computational paradigm has emerged in computer both Renault and Toyota have deployed online configuration science over the past few decades, which is exemplified by systems based on knowledge compilation). QBF solvers the use of SAT solvers to tackle problems in the complexity have been used in model checking, verification, debugging, class NP. Finally, function problem solvers have and engineering investment is made towards developing been used in model-based diagnosis, design debugging, highly efficient solvers for a prototypical problem CAD and bioinformatics. The cost of this investment is then on a variety of topics, including algorithms; descriptions amortized as these solvers are applied to a broader class of of implementations and/or evaluations of beyond NP problems via reductions (in contrast to developing dedicated solvers; their applications (including encodings); the complexity algorithms for each encountered problem). SAT solvers, classes they reach; and their connections to one for example, are now routinely used to solve problems in another.

Decision tree classifiers are a widely used tool in data stream mining. The use of confidence intervals to estimate the gain associated with each split leads to very effective methods, like the popular Hoeffding tree algorithm. From a statistical viewpoint, the analysis of decision tree classifiers in a streaming setting requires knowing when enough new information has been collected to justify splitting a leaf. Although some of the issues in the statistical analysis of Hoeffding trees have been already clarified, a general and rigorous study of confidence intervals for splitting criteria is missing. We fill this gap by deriving accurate confidence intervals to estimate the splitting gain in decision tree learning with respect to three criteria: entropy, Gini index, and a third index proposed by Kearns and Mansour. Our confidence intervals depend in a more detailed way on the tree parameters. We also extend our confidence analysis to a selective sampling setting, in which the decision tree learner adaptively decides which labels to query in the stream. We furnish theoretical guarantee bounding the probability that the classification is non-optimal learning the decision tree via our selective sampling strategy. Experiments on real and synthetic data in a streaming setting show that our trees are indeed more accurate than trees with the same number of leaves generated by other techniques and our active learning module permits to save labeling cost. In addition, comparing our labeling strategy with recent methods, we show that our approach is more robust and consistent respect all the other techniques applied to incremental decision trees.