The concept of stochastic configuration networks (SCNs) others a solid framework for fast implementation of feedforward neural networks through randomized learning. Unlike conventional randomized approaches, SCNs provide an avenue to select appropriate scope of random parameters to ensure the universal approximation property. In this paper, a deep version of stochastic configuration networks, namely deep stacked stochastic configuration network (DSSCN), is proposed for modeling non-stationary data streams. As an extension of evolving stochastic connfiguration networks (eSCNs), this work contributes a way to grow and shrink the structure of deep stochastic configuration networks autonomously from data streams. The performance of DSSCN is evaluated by six benchmark datasets. Simulation results, compared with prominent data stream algorithms, show that the proposed method is capable of achieving comparable accuracy and evolving compact and parsimonious deep stacked network architecture.

When learning from positive and unlabelled data, it is a strong assumption that the positive observations are randomly sampled from the distribution of $X$ conditional on $Y = 1$, where X stands for the feature and Y the label. Most existing algorithms are optimally designed under the assumption. However, for many real-world applications, the observed positive examples are dependent on the conditional probability $P(Y = 1|X)$ and should be sampled biasedly. In this paper, we assume that a positive example with a higher $P(Y = 1|X)$ is more likely to be labelled and propose a probabilistic-gap based PU learning algorithms. Specifically, by treating the unlabelled data as noisy negative examples, we could automatically label a group positive and negative examples whose labels are identical to the ones assigned by a Bayesian optimal classifier with a consistency guarantee. The relabelled examples have a biased domain, which is remedied by the kernel mean matching technique. The proposed algorithm is model-free and thus do not have any parameters to tune. Experimental results demonstrate that our method works well on both generated and real-world datasets.

Mastering the dynamics of social influence requires separating, in a database of information propagation traces, the genuine causal processes from temporal correlation, homophily and other spurious causes. However, most of the studies to characterize social influence and, in general, most data-science analyses focus on correlations, statistical independence, conditional independence etc.; only recently, there has been a resurgence of interest in "causal data science", e.g., grounded on causality theories. In this paper we adopt a principled causal approach to the analysis of social influence from information-propagation data, rooted in probabilistic causal theory. Our approach develops around two phases. In the first step, in order to avoid the pitfalls of misinterpreting causation when the data spans a mixture of several subtypes ("Simpson's paradox"), we partition the set of propagation traces in groups, in such a way that each group is as less contradictory as possible in terms of the hierarchical structure of information propagation. For this goal we borrow from the literature the notion of "agony" and define the Agony-bounded Partitioning problem, which we prove being hard, and for which we develop two efficient algorithms with approximation guarantees. In the second step, for each group from the first phase, we apply a constrained MLE approach to ultimately learn a minimal causal topology. Experiments on synthetic data show that our method is able to retrieve the genuine causal arcs w.r.t. a known ground-truth generative model. Experiments on real data show that, by focusing only on the extracted causal structures instead of the whole social network, we can improve the effectiveness of predicting influence spread.

We develop unbiased implicit variational inference (UIVI), a method that expands the applicability of variational inference by defining an expressive variational family. UIVI considers an implicit variational distribution obtained in a hierarchical manner using a simple reparameterizable distribution whose variational parameters are defined by arbitrarily flexible deep neural networks. Unlike previous works, UIVI directly optimizes the evidence lower bound (ELBO) rather than an approximation to the ELBO. We demonstrate UIVI on several models, including Bayesian multinomial logistic regression and variational autoencoders, and show that UIVI achieves both tighter ELBO and better predictive performance than existing approaches at a similar computational cost.

In several real-world applications involving decision making under uncertainty, the traditional expected value objective may not be suitable, as it may be necessary to control losses in the case of a rare but extreme event. Conditional Value-at-Risk (CVaR) is a popular risk measure for modeling the aforementioned objective. We consider the problem of estimating CVaR from i.i.d. samples of an unbounded random variable, which is either sub-Gaussian or sub-exponential. We derive a novel one-sided concentration bound for a natural sample-based CVaR estimator in this setting. Our bound relies on a concentration result for a quantile-based estimator for Value-at-Risk (VaR), which may be of independent interest.

This paper describes a novel perspective on the foundations of mathematics: how mathematics may be seen to be largely about 'information compression via the matching and unification of patterns' (ICMUP). ICMUP is itself a novel approach to information compression, couched in terms of non-mathematical primitives, as is necessary in any investigation of the foundations of mathematics. This new perspective on the foundations of mathematics has grown out of an extensive programme of research developing the "SP Theory of Intelligence" and its realisation in the "SP Computer Model", a system in which a generalised version of ICMUP -- the powerful concept of SP-multiple-alignment -- plays a central role. These ideas may be seen to be part of a "Big Picture" comprising six areas of interest, with information compression as a unifying theme. The paper describes the close relation between mathematics and information compression, and describes examples showing how variants of ICMUP may be seen in widely-used structures and operations in mathematics. Examples are also given to show how the mathematics-related disciplines of logic and computing may be understood as ICMUP. There are many potential benefits and applications of these ideas.

Structural causal models (SCMs), also known as (non-parametric) structural equation models (SEMs), are widely used for causal modeling purposes. A large body of theoretical results is available for the special case in which cycles are absent (i.e., acyclic SCMs, also known as recursive SEMs). However, in many application domains cycles are abundantly present, for example in the form of feedback loops. In this paper, we provide a general and rigorous theory of cyclic SCMs. The paper consists of two parts: the first part gives a rigorous treatment of structural causal models, dealing with measure-theoretic and other complications that arise in the presence of cycles. In contrast with the acyclic case, in cyclic SCMs solutions may no longer exist, or if they exist, they may no longer be unique, or even measurable in general. We give several sufficient and necessary conditions for the existence of (unique) measurable solutions. We show how causal reasoning proceeds in these models and how this differs from the acyclic case. Moreover, we give an overview of the Markov properties that hold for cyclic SCMs. In the second part, we address the question of how one can marginalize an SCM (possibly with cycles) to a subset of the endogenous variables. We show that under a certain condition, one can effectively remove a subset of the endogenous variables from the model, leading to a more parsimonious marginal SCM that preserves the causal and counterfactual semantics of the original SCM on the remaining variables. Moreover, we show how the marginalization relates to the latent projection and to latent confounders, i.e. latent common causes.

Bayesian Networks have been widely used in the last decades in many fields, to describe statistical dependencies among random variables. In general, learning the structure of such models is a problem with considerable theoretical interest that still poses many challenges. On the one hand, this is a well-known NP-complete problem, which is practically hardened by the huge search space of possible solutions. On the other hand, the phenomenon of I-equivalence, i.e., different graphical structures underpinning the same set of statistical dependencies, may lead to multimodal fitness landscapes further hindering maximum likelihood approaches to solve the task. Despite all these difficulties, greedy search methods based on a likelihood score coupled with a regularization term to account for model complexity, have been shown to be surprisingly effective in practice. In this paper, we consider the formulation of the task of learning the structure of Bayesian Networks as an optimization problem based on a likelihood score. Nevertheless, our approach do not adjust this score by means of any of the complexity terms proposed in the literature; instead, it accounts directly for the complexity of the discovered solutions by exploiting a multi-objective optimization procedure. To this extent, we adopt NSGA-II and define the first objective function to be the likelihood of a solution and the second to be the number of selected arcs. We thoroughly analyze the behavior of our method on a wide set of simulated data, and we discuss the performance considering the goodness of the inferred solutions both in terms of their objective functions and with respect to the retrieved structure. Our results show that NSGA-II can converge to solutions characterized by better likelihood and less arcs than classic approaches, although paradoxically frequently characterized by a lower similarity to the target network.

Bayesian network modelling is a well adapted approach to study messy and highly correlated datasets which are very common in, e.g., systems epidemiology. A popular approach to learn a Bayesian network from an observational datasets is to identify the maximum a posteriori network in a search-and-score approach. Many scores have been proposed both Bayesian or frequentist based. In an applied perspective, a suitable approach would allow multiple distributions for the data and is robust enough to run autonomously. A promising framework to compute scores are generalized linear models. Indeed, there exists fast algorithms for estimation and many tailored solutions to common epidemiological issues. The purpose of this paper is to present an R package abn that has an implementation of multiple frequentist scores and some realistic simulations that show its usability and performance. It includes features to deal efficiently with data separation and adjustment which are very common in systems epidemiology.

A novel text-independent speaker identification (SI) method is proposed. This method uses the Mel-frequency Cepstral coefficients (MFCCs) and the dynamic information among adjacent frames as feature sets to capture speaker's characteristics. In order to utilize dynamic information, we design super-MFCCs features by cascading three neighboring MFCCs frames together. The probability density function (PDF) of these super-MFCCs features is estimated by the recently proposed histogram transform~(HT) method, which generates more training data by random transforms to realize the histogram PDF estimation and recedes the commonly occurred discontinuity problem in multivariate histograms computing. Compared to the conventional PDF estimation methods, such as Gaussian mixture models, the HT model shows promising improvement in the SI performance.