A very recent algorithm shows search of games of imperfect information has been around how both theoretical problems can be fixed (Lisý, Lanctot, for many years. The approach is appealing, for a number of and Bowling 2015), but has yet to be applied to large games reasons: it allows the usage of well-known methods from typically used for search. More recently overestimation of perfect information games, its complexity is magnitudes MAX's knowledge is also dealt with in the field of general lower than the problem of weakly solving a game in the game play (Schofield, Cerexhe, and Thielscher 2013). To the sense of game theory, it can be used in a justin-time manner best of our knowledge, all literature on the deficiencies of (no precalculation phase needed) even for games with PIMC concentrates on the overestimation of MAX's knowledge.

Wildlife poaching presents a serious extinction threat to many animal species. In order to save wildlife in designated wildlife parks, park rangers conduct patrols over the park area to combat such illegal activities. An important aspect of the patrolling activity of the rangers is to anticipate where the poachers are likely to catch animals and then respond accordingly. Previous work has applied defender-attacker Stackelberg Security Games (SSGs) to solve the problem of wildlife protection, wherein attacker behavioral models are used to predict the behaviors of the poachers. However, these behavioral models have several limitations which limit their accuracy in predicting poachers' behavior. First, existing models fail to account for the rangers' imperfect observations w.r.t poaching activities (due to the limited capability of rangers to patrol thoroughly over a vast geographical area). Second, these models are built upon discrete choice models that assume a single agent choosing targets, while it is infeasible to obtain information about every single attacker in wildlife protection. Third, these models do not consider the effect of past poachers' actions on the current poachers' activities, one of the key factors affecting the poachers' behaviors. In this work, we attempt to address these limitations while providing three main contributions. First, we propose a novel hierarchical behavioral model, HiBRID, to predict the poachers' behaviors wherein the rangers' imperfect detection of poaching signs is taken into account --- a significant advance towards existing behavioral models in security games. Furthermore, HiBRID incorporates the temporal effect on the poachers' behaviors. The model also does not require a known number of attackers. Second, we provide two new heuristics: \textit{parameter separation} and \textit{target abstraction} to reduce the computational complexity in learning the model parameters. Finally, we use the real-world data collected in Queen Elizabeth National Park (QENP) in Uganda over 12 years to evaluate the prediction accuracy of our new model.

Over the last decade, building on advances in the areas of knowledge compilation and weighted model counting has drastically increased the scalability of inference in probabilistic logic programs. In this paper, we provide an overview of how this has been possible and point out some open challenges.

This paper overviews recent work on compilation of relational queries into lossless factorized representations. The primary motivation for this compilation is to avoid redundancy in the representation of query results and speed up their computation and subsequent analytics.

The role of decomposition-trees (also known as junction and clique trees) in probabilistic inference is widely known and has been the basis for many well known inference algorithms.Recent approaches have demonstrated that such trees have a "hidden structure," which enables the characterization of tractable problem instances as well as lead to insights that enable boosting the performance of inference algorithms. We consider the MPE problem on a Boolean formula in CNF where each literal in the formula is associated with a weight.We describe techniques for exploiting the junction-tree structure of these formulas in the context of a branch-and-bound algorithm for MPE.

Compiling a Bayesian network into a secondary structure, such as a junction tree or arithmetic circuit allows for offline computations before observations arrive, and quick inference for the marginal of all variables. However, query-based algorithms, such as variable elimination and recursive conditioning, that compute the posterior marginal of few variables given some observations, allow pruning of irrelevant variables, which can reduce the size of the problem. Madsen and Jensen show how lazy evaluation of junction trees can allow both compilation and pruning. In this paper, we adapt the lazy evaluation to arithmetic circuits, allowing the best of both worlds: pruning due to observations and query variables as well as compilation while exploiting local structure and determinism.

SAT and #SAT are at the heart of many important problem formulations in AI, the most prominent being reasoning and learning in first-order and probabilistic knowledge bases. In practice, all contemporary systems resort to domain closure: objects in the universe are all and only the ones mentioned in the knowledge base. This is in stark contrast to the natural ability of human beings to infer things about sensory inputs and unforeseen data: they infer the existence of objects from their observations; no predefined list of objects is given or known in advance. In this paper, we introduce the formal foundations for reasoning in open universes in a general way, purely based on SAT and #SAT technology.

In this paper an extension to factorial hidden Semi Markov Models is introduced that allows modeling more than one sequence of emissions of the individual HMM chains, as well as a joint emission of all chains. Since exact inference in factorial hidden Markov Models is computationally intractable, an approximate inference technique is introduced that reduces the computational costs by first constraining the successor state space of the model, allowing state changes at statistically significant points in time (events) and by discarding low probability paths (truncating). Furthermore, by being agnostic about state durations the computational costs are further decreased. These assumptions allow for efficient inference that is less susceptible to local minima and allows one to specify the computational burden a priori. The performance of the inference technique is evaluated empirically on a synthetic data set whereas incorporating the feature emissions is evaluated on real world data in the context of energy disaggregation. Energy disaggregation tackles the problem of decomposing whole home energy measurements into the power traces of constituent appliances, and is a natural application for this type of models.

Selecting homes with energy-efficient infrastructure is important for renters, because infrastructure influences energy consumption more than in-home behavior.Personalized energy estimates can guide prospective tenants toward energy-efficient homes, but this information is not readily available. Utility estimates are not typically offered to house-hunters, and existing technologies like carbon calculators require users to answer (prohibitively) many questions that may require considerable research to answer. For the task of providing personalized utility estimates to prospective tenants, we present a cost-based model for feature selection at training time, where all features are available and costs assigned to each feature reflect the difficulty of acquisition. At test time, we have immediate access to some features but others are difficult to acquire (costly). In this limited-information setting, we strategically order questions we ask each user, tailored to previous information provided, to give the most accurate predictions while minimizing the cost to users. During the critical first 10 questions that our approach selects, prediction accuracy improves equally to fixed order approaches, but prediction certainty is higher.

While an immense amount of work has gone into novel clustering algorithms, little work has focused on developing compact, domain-specific explanations for the results of the clustering algorithms. Attaching semantic meaning to a cluster has numerous benefits, including the ability for such a description to be both human and machine readable. In this paper, we assume that the clusters are given to us, and find the minimal set of features that can differentiate one cluster from the remaining set of samples. We formulate this problem as an integer linear program. By using samples not belonging to the cluster in the optimization formulation, the resulting description will be minimal and contain no false positives. The efficacy of this method is demonstrated on simulation data and real-world malware data run in a sandbox that collects behavioral characteristics. In the case of malware, once it has been clustered, it would have been sent to a reverse engineer who would have been tasked with creating the actual meaning of the clustering results and disseminating this information through signatures or indicators of compromise. This is a time-consuming process that can take hours to weeks depending on the complexity of the malware family. The methods presented in this paper automatically generate optimal signatures, which can then be quickly propagated to help contain the spread of a malware family.