Networks are ubiquitous in science and have become a focal point for discussion in everyday life. Formal statistical models for the analysis of network data have emerged as a major topic of interest in diverse areas of study, and most of these involve a form of graphical representation. Probability models on graphs date back to 1959. Along with empirical studies in social psychology and sociology from the 1960s, these early works generated an active network community and a substantial literature in the 1970s. This effort moved into the statistical literature in the late 1970s and 1980s, and the past decade has seen a burgeoning network literature in statistical physics and computer science. The growth of the World Wide Web and the emergence of online networking communities such as Facebook, MySpace, and LinkedIn, and a host of more specialized professional network communities has intensified interest in the study of networks and network data. Our goal in this review is to provide the reader with an entry point to this burgeoning literature. We begin with an overview of the historical development of statistical network modeling and then we introduce a number of examples that have been studied in the network literature. Our subsequent discussion focuses on a number of prominent static and dynamic network models and their interconnections. We emphasize formal model descriptions, and pay special attention to the interpretation of parameters and their estimation. We end with a description of some open problems and challenges for machine learning and statistics.

The problem is sequence prediction in the following setting. A sequence $x_1,...,x_n,...$ of discrete-valued observations is generated according to some unknown probabilistic law (measure) $\mu$. After observing each outcome, it is required to give the conditional probabilities of the next observation. The measure $\mu$ belongs to an arbitrary but known class $C$ of stochastic process measures. We are interested in predictors $\rho$ whose conditional probabilities converge (in some sense) to the "true" $\mu$-conditional probabilities if any $\mu\in C$ is chosen to generate the sequence. The contribution of this work is in characterizing the families $C$ for which such predictors exist, and in providing a specific and simple form in which to look for a solution. We show that if any predictor works, then there exists a Bayesian predictor, whose prior is discrete, and which works too. We also find several sufficient and necessary conditions for the existence of a predictor, in terms of topological characterizations of the family $C$, as well as in terms of local behaviour of the measures in $C$, which in some cases lead to procedures for constructing such predictors. It should be emphasized that the framework is completely general: the stochastic processes considered are not required to be i.i.d., stationary, or to belong to any parametric or countable family.

Interest in multioutput kernel methods is increasing, whether under the guise of multitask learning, multisensor networks or structured output data. From the Gaussian process perspective a multioutput Mercer kernel is a covariance function over correlated output functions. One way of constructing such kernels is based on convolution processes (CP). A key problem for this approach is efficient inference. Alvarez and Lawrence (2009) recently presented a sparse approximation for CPs that enabled efficient inference. In this paper, we extend this work in two directions: we introduce the concept of variational inducing functions to handle potential non-smooth functions involved in the kernel CP construction and we consider an alternative approach to approximate inference based on variational methods, extending the work by Titsias (2009) to the multiple output case. We demonstrate our approaches on prediction of school marks, compiler performance and financial time series.

A central problem in artificial intelligence is that of planning to maximize future reward under uncertainty in a partially observable environment. In this paper we propose and demonstrate a novel algorithm which accurately learns a model of such an environment directly from sequences of action-observation pairs. We then close the loop from observations to actions by planning in the learned model and recovering a policy which is near-optimal in the original environment. Specifically, we present an efficient and statistically consistent spectral algorithm for learning the parameters of a Predictive State Representation (PSR). We demonstrate the algorithm by learning a model of a simulated high-dimensional, vision-based mobile robot planning task, and then perform approximate point-based planning in the learned PSR. Analysis of our results shows that the algorithm learns a state space which efficiently captures the essential features of the environment. This representation allows accurate prediction with a small number of parameters, and enables successful and efficient planning.

The development of tools to provide insight into the behavioral response of a civilian population will greatly benefit the modeling and simulation community and have potential applications across multiple user communities in the U.S. Department of Defense. We present an overview of a modular agent-based modeling framework, grounded in the human behavioral and social theory, which is intended to represent a populations’ stance on issues as a function of their changing beliefs, values and interests. We utilize and integrate theories of narrative identity [1] and planned behavior [2] with macrosociological theories of heterogeneity and influence [3][4] to model civilian behavior in a conflict ecosystem. Communication between agents takes place across a social network developed using real data about the population under consideration, and essential services are implemented as objects within the model allowing for experimentation with different courses of action for development of civil service capacity. We describe the theoretical underpinnings of the model, the current state of implementation, potential use cases, and the path forward for future work.

After building a classifier with modern tools of machine learning we typically have a black box at hand that is able to predict well for unseen data. Thus, we get an answer to the question what is the most likely label of a given unseen data point. However, most methods will provide no answer why the model predicted the particular label for a single instance and what features were most influential for that particular instance. The only method that is currently able to provide such explanations are decision trees. This paper proposes a procedure which (based on a set of assumptions) allows to explain the decisions of any classification method.

Some alternatives to probability in the literature include sets of probability measure [Huber 1981; Walley 1991], Dempster-Shafer belief functions [Shafer 1976] and the closely related nonadditive measures [Schmeidler 1989], and possibility measures [Dubois and Prade 1990]. In this paper, we consider the notion of expectation for all these representations of uncertainty. We do not take a stand here on what the "right" way is to represent uncertainty; we simply investigate characterizations of expectation and reasoning about expectation, both for probability and for other representations of uncertainty. It is well known that a probability measure determines a unique expectation function that is linear (i.e., E (aX + bY) = aE (X) + bE (Y)), monotone (i.e., X Y implies E ( X) E (Y)), and maps constant functions to their value. Conversely, given an expectation function E (that is, a function from random variables to the reals) that is linear, monotone, and maps constant functions to their value, there is a unique probability measure µ such that E = E µ. That is, there is a 1-1 mapping from probability measures to (probabilistic) expectation functions. One of the goals of this paper is to provide similar characterizations of expectation for other representations of uncertainty. Some work along these lines has already been done, particulary with regard to sets of probability measures [Huber 1981; Walley 1991; 1981]. 1 However, there seems to be surprisingly little work on characterizing expectation in the context of other measures of uncertainty, such as belief functions [Shafer 1976] and possibility measures [Dubois and Prade 1990].

Probability-like parameters appearing in some statistical models, and their prior distributions, are reinterpreted through the notion of `circumstance', a term which stands for any piece of knowledge that is useful in assigning a probability and that satisfies some additional logical properties. The idea, which can be traced to Laplace and Jaynes, is that the usual inferential reasonings about the probability-like parameters of a statistical model can be conceived as reasonings about equivalence classes of `circumstances' - viz., real or hypothetical pieces of knowledge, like e.g. physical hypotheses, that are useful in assigning a probability and satisfy some additional logical properties - that are uniquely indexed by the probability distributions they lead to.

An approach to induction is presented, based on the idea of analysing the context of a given problem into `circumstances'. This approach, fully Bayesian in form and meaning, provides a complement or in some cases an alternative to that based on de Finetti's representation theorem and on the notion of infinite exchangeability. In particular, it gives an alternative interpretation of those formulae that apparently involve `unknown probabilities' or `propensities'. Various advantages and applications of the presented approach are discussed, especially in comparison to that based on exchangeability. Generalisations are also discussed.

The multi-armed bandit is a concise model for the problem of iterated decision-making under uncertainty. In each round, a gambler must pull one of $K$ arms of a slot machine, without any foreknowledge of their payouts, except that they are uniformly bounded. A standard objective is to minimize the gambler's regret, defined as the gambler's total payout minus the largest payout which would have been achieved by any fixed arm, in hindsight. Note that the gambler is only told the payout for the arm actually chosen, not for the unchosen arms. Almost all previous work on this problem assumed the payouts to be non-adaptive, in the sense that the distribution of the payout of arm $j$ in round $i$ is completely independent of the choices made by the gambler on rounds $1, \dots, i-1$. In the more general model of adaptive payouts, the payouts in round $i$ may depend arbitrarily on the history of past choices made by the algorithm. We present a new algorithm for this problem, and prove nearly optimal guarantees for the regret against both non-adaptive and adaptive adversaries. After $T$ rounds, our algorithm has regret $O(\sqrt{T})$ with high probability (the tail probability decays exponentially). This dependence on $T$ is best possible, and matches that of the full-information version of the problem, in which the gambler is told the payouts for all $K$ arms after each round. Previously, even for non-adaptive payouts, the best high-probability bounds known were $O(T^{2/3})$, due to Auer, Cesa-Bianchi, Freund and Schapire. The expected regret of their algorithm is $O(T^{1/2}) for non-adaptive payouts, but as we show, $Ω(T^{2/3})$ for adaptive payouts.