We present a probabilistic model of an intrusion in a marked renewal process. Given a process and a sequence of events, an intrusion is a subsequence of events that is not produced by the process. Applications of the model are, for example, online payment fraud with the fraudster taking over a user's account and performing payments on the user's behalf, or unexpected equipment failures due to unintended use. We adopt Bayesian approach to infer the probability of an intrusion in a sequence of events, a MAP subsequence of events constituting the intrusion, and the marginal probability of each event in a sequence to belong to the intrusion. We evaluate the model for intrusion detection on synthetic data, as well as on anonymized data from an online payment system.

Modern neural networks are highly overparameterized, with capacity to substantially overfit to training data. Nevertheless, these networks often generalize well in practice. It has also been observed that trained networks can often be "compressed" to much smaller representations. The purpose of this paper is to connect these two empirical observations. Our main technical result is a generalization bound for compressed networks based on the compressed size. Combined with off-the-shelf compression algorithms, the bound leads to state of the art generalization guarantees; in particular, we provide the first non-vacuous generalization guarantees for realistic architectures applied to the ImageNet classification problem. As additional evidence connecting compression and generalization, we show that compressibility of models that tend to overfit is limited: We establish an absolute limit on expected compressibility as a function of expected generalization error, where the expectations are over the random choice of training examples. The bounds are complemented by empirical results that show an increase in overfitting implies an increase in the number of bits required to describe a trained network.

A common mechanism to assess trust in crowdworkers is to have them answer gold tasks. However, assigning gold tasks to all workers reduces the efficiency of the platform. We propose a mechanism that exploits transitivity so that a worker can be certified as trusted by other trusted workers who solve common tasks. Thus, trust can be derived from a smaller number of gold tasks assignment through multiple layers of peer relationship among the workers, a model we call deep trust. We use the derived trust to incentivize workers for high quality work and show that the resulting mechanism is dominant strategy incentive compatible. We also show that the mechanism satisfies a notion of fairness in that the trust assessment (and thus the reward) of a worker in the limit is independent of the quality of other workers.

Bayesian approaches have been used extensively in scientific and engineering research to quantify uncertainty and extract information. However, its model-dependent nature means that when the a priori model is incomplete or unavailable, there is a severe risk that Bayesian approaches will yield misleading results. Here, we propose a universal model-free information extraction approach, capable of reliably recovering target signals from complex responses. This breakthrough leverages on a data-centric approach, whereby measured data is reconfigured to create an enriched observable space, which in turn is mapped to a well-adapted manifold, thereby detecting crucial information via a reconstructed low-rank phase-space. A Koopman operator is used to transform hidden and complex nonlinear dynamics to linear one, which enables us to detect hidden event of interest from rapidly evolving systems, and relate it to either unobservable stimulus or anomalous behaviour. Thanks to its data-driven nature, our method excludes completely any prior knowledge on governing dynamics. We benchmark the astonishing accuracy of our method on three diverse and challenging problems in: biology, medicine, and engineering. In all cases, our approach outperforms existing state-of-the-art methods, of both Bayesian and non-Bayesian type. By creating a new reliable information analysis paradigm, it is suitable for ubiquitous nonlinear dynamical systems or end-users with little expertise, which permits the unbiased understanding of various mechanisms in the real world.

The fundamental problem in short-text classification is \emph{feature sparseness} -- the lack of feature overlap between a trained model and a test instance to be classified. We propose \emph{ClassiNet} -- a network of classifiers trained for predicting missing features in a given instance, to overcome the feature sparseness problem. Using a set of unlabeled training instances, we first learn binary classifiers as feature predictors for predicting whether a particular feature occurs in a given instance. Next, each feature predictor is represented as a vertex $v_i$ in the ClassiNet where a one-to-one correspondence exists between feature predictors and vertices. The weight of the directed edge $e_{ij}$ connecting a vertex $v_i$ to a vertex $v_j$ represents the conditional probability that given $v_i$ exists in an instance, $v_j$ also exists in the same instance. We show that ClassiNets generalize word co-occurrence graphs by considering implicit co-occurrences between features. We extract numerous features from the trained ClassiNet to overcome feature sparseness. In particular, for a given instance $\vec{x}$, we find similar features from ClassiNet that did not appear in $\vec{x}$, and append those features in the representation of $\vec{x}$. Moreover, we propose a method based on graph propagation to find features that are indirectly related to a given short-text. We evaluate ClassiNets on several benchmark datasets for short-text classification. Our experimental results show that by using ClassiNet, we can statistically significantly improve the accuracy in short-text classification tasks, without having to use any external resources such as thesauri for finding related features.

The most recent financial upheavals have cast doubt on the adequacy of some of the conventional quantitative risk management strategies, such as VaR (Value at Risk), in many common situations. Consequently, there has been an increasing need for verisimilar financial stress testings, namely simulating and analyzing financial portfolios in extreme, albeit rare scenarios. Unlike conventional risk management which exploits statistical correlations among financial instruments, here we focus our analysis on the notion of probabilistic causation, which is embodied by Suppes-Bayes Causal Networks (SBCNs); SBCNs are probabilistic graphical models that have many attractive features in terms of more accurate causal analysis for generating financial stress scenarios. In this paper, we present a novel approach for conducting stress testing of financial portfolios based on SBCNs in combination with classical machine learning classification tools. The resulting method is shown to be capable of correctly discovering the causal relationships among financial factors that affect the portfolios and thus, simulating stress testing scenarios with a higher accuracy and lower computational complexity than conventional Monte Carlo Simulations.

One of the critical issues when adopting Bayesian networks (BNs) to model dependencies among random variables is to "learn" their structure. This is a well-known NP-hard problem in its most general and classical formulation, which is furthermore complicated by known pitfalls such as the issue of I-equivalence among different structures. In this work we restrict the investigation to a specific class of networks, i.e., those representing the dynamics of phenomena characterized by the monotonic accumulation of events. Such phenomena allow to set specific structural constraints based on Suppes' theory of probabilistic causation and, accordingly, to define constrained BNs, named Suppes-Bayes Causal Networks (SBCNs). Within this framework, we study the structure learning of SBCNs via extensive simulations with various state-of-the-art search strategies, such as canonical local search techniques and Genetic Algorithms. This investigation is intended to be an extension and an in-depth clarification of our previous works on SBCN structure learning. Among the main results, we show that Suppes' constraints do simplify the learning task, by reducing the solution search space and providing a temporal ordering on the variables, which simplifies the complications derived by I-equivalent structures. Finally, we report on tradeoffs among different optimization techniques that can be used to learn SBCNs.

This article describes an implementation of a nonparametric Bayesian approach to solving binary classification problems on graphs. We consider a hierarchical Bayesian approach with a prior that is constructed by truncating a series expansion of the soft label function using the graph Laplacian eigenfunctions as basis functions. We compare our truncated prior to the untruncated Laplacian based prior in simulated and real data examples to illustrate the improved scalability in terms of size of the underlying graph.

Finite mixture models have become a popular tool for clustering. Amongst other uses, they have been applied for clustering longitudinal data and clustering high-dimensional data. In the latter case, a latent Gaussian mixture model is sometimes used. Although there has been much work on clustering using latent variables and on clustering longitudinal data, respectively, there has been a paucity of work that combines these features. An approach is developed for clustering longitudinal data with many time points based on an extension of the mixture of common factor analyzers model. A variation of the expectation-maximization algorithm is used for parameter estimation and the Bayesian information criterion is used for model selection. The approach is illustrated using real and simulated data.

Discovering the causal structure among a set of variables is a fundamental problem in many areas of science. In this paper, we propose Kernel Conditional Deviance for Causal Inference (KCDC) a fully nonparametric causal discovery method based on purely observational data. From a novel interpretation of the notion of asymmetry between cause and effect, we derive a corresponding asymmetry measure using the framework of reproducing kernel Hilbert spaces. Based on this, we propose three decision rules for causal discovery. We demonstrate the wide applicability of our method across a range of diverse synthetic datasets. Furthermore, we test our method on real-world time series data and the real-world benchmark dataset Tubingen Cause-Effect Pairs where we outperform existing state-of-the-art methods.