Data centers today are heterogeneous: they have servers from multiple generations and multiple vendors; server machines have multiple cores that are capable of running at difference speeds, and some have general purpose graphics processing units (GPGPU). Hardware trends indicate that future processors will have heterogeneous cores with different speeds and capabilities. This environment enables new advances in power saving and application optimization. It also poses new challenges, as current systems software is ill-suited for heterogeneity. In this position paper, we focus on interactive applications and outline some of the techniques to embrace heterogeneity. We show that heterogeneity can be exploited to deliver interactive services in an energy-efficient manner. For example, our initial study suggests that neither high-end nor low-end servers alone are very effective in servicing a realistic workload, which typically has requests with varying service demands. High-end servers achieve good throughput but the energy costs are high. Low-end servers are energy-efficient for short requests, but they may not be able to serve long requests at the desired quality of service. In this work, we show that a heterogeneous system can be a better choice than an equivalent homogeneous system to deliver interactive services in a cost-effective manner, transforming heterogeneity from a resource management nightmare to an asset. We highlight some of the challenges and opportunities and the role of AI and machine learning techniques for hosting large interactive services in data centers.

Despite a dramatic growth of power consumption inhouseholds, less attention has been paid to monitoring,analyzing and predicting energy usage. In this paper,we propose a framework to mine raw energy data bytransforming time series energy data into a symbol se-quence, and then extend a suffix tree data structure asan efficient representation to analyze global structuralpatterns. Then, we use a clustering algorithm to detectenergy pattern outliers which are far from their clustercentroids. To validate our approach, we use real powerdata collected from a smart apartment testbed duringtwo months.

Adversarial behavioral has been observed in many different contexts. In this paper we address the problem of adversarial behavior in the context of clandestine networks. We use data from a massively multiplayer online role playing game to illustrate the behavioral and structural signatures of deviant players change over time as a response to "policing" activities of the game administrators. Preliminary results show that the behavior of the deviant players and their affiliates show co-evolutionary behavior and the timespan within the game can be divided into different epochs based on their behaviors. Feature sets derived from these results can be used for better predictive machine learning models for detecting deviants in clandestine networks.

The goal for attackers is to move from one side of the graph to the Border security is a key element of national security policy other (represented by sets of source and target nodes); this for any sovereign nation. In the United States, the Border represents a typical scenario of crossing an open region from Patrol deploys thousands of agents integrated with technology one side of the border to destination points in the interior of (e.g., vehicles, cameras, sensors) and infrastructure the county. The paths between the source and target nodes (e.g., fences, checkpoints) to prevent illegal entry of people may represent major or minor roads, or paths suitable for and goods into the country along vast land borders with travel on foot. We use weights on the edges to represent Canada and Mexico. The problem of border security is incredibly the relative speed/cost of transit on the different paths (for complex, due to the diversity and volume of illegal example, it may be must slower and more difficult to use activity that must be controlled, the variety of resources that a foot path than a major highway). Nodes may represent can be deployed to secure the border, and the differences in intersections, checkpoints, or other important waypoints.

Game theory has become an important tools for making resource allocations decision in security domains, including critical infrastructure protection. Many of these games are formulated as Stackelberg security games. We present new analysis and algorithms for a class of Stackelberg security games with identical, fully protective defender resources. The first algorithm has worst-case complexity linear in the number of possible targets, but works only for a restricted case. The second algorithm can find and optimal resource allocation for the general case in time O(n log(n)).

We introduce the notion of strategic uncertainty for boundedly rational, non-myopic agents as an analog to the equilibrium selection problem in classical game theory. We then motivate the need for and feasibility of addressing strategic uncertainty and present an algorithm that produces decisions that are robust to it. Finally, we show how agents' rationality levels and planning horizons alter the robustness of their decisions.

In this paper, we introduce new capabilities to our existing microsimulation framework, Simulacron. These new capabilities add the modelling of behaviours based on motivations and improve our existing non-deterministic movement capacity. We then discuss the application of these new features to a simple, synthetic, proof of concept, scenario involving the transit of people through a corridor and how an induced panic affects their throughput. Finally we describe a more complex scenario, which is currently under development, involving the detonation of an explosive device in a major metropolitan transport hub at peak hour and the analysis of subsequent reaction.

Speech acts provide good insights into the communicative behavior of tweeters on Twitter. This paper is mainly concerned with speech act recognition in Twitter as a multi-class classification problem, for which we propose a set of word-based and character-based features. Inexpensive, robust and efficient, our method achieves an average F1 score of nearly 0.7 with the existence of much noise in our annotated Twitter data. In view of the deficiency of training data for the task, we experimented extensively with different configurations of training and test data, leading to empirical findings that may provide valuable reference for building benchmark datasets for sustained research on speech act recognition in Twitter.

Automatic detection of persuasion is essential for machine interaction on the social web. To facilitate automated persuasion detection, we present a novel microtext corpus derived from hostage negotiation transcripts as well as a detailed manual (codebook) for persuasion annotation. Our corpus, called the NPS Persuasion Corpus, consists of 37 transcripts from four sets of hostage negotiation transcriptions. Each utterance in the corpus is hand annotated for one of nine categories of persuasion based on Cialdini’s model: reciprocity, commitment, consistency, liking, authority, social proof, scarcity, other, and not persuasive. Initial results using three supervised learning algorithms (Na ̈ve Bayes, Maximum Entropy, and Support Vector Machines) combined with gappy and orthogonal sparse bigram feature expansion techniques show that the annotation process did capture machine learnable features of persuasion with F-scores better than baseline.

We evaluate the statistical dependency parser, Malt, on a new dataset of sentences taken from tweets. We use a version of Malt which is trained on gold standard phrase structure Wall Street Journal (WSJ) trees converted to Stanford labelled dependencies. We observe a drastic drop in performance moving from our in-domain WSJ test set to the new Twitter dataset, much of which has to do with the propagation of part-of-speech tagging errors. Retraining Malt on dependency trees produced by a state-of-the-art phrase structure parser, which has itself been self-trained on Twitter material, results in a significant improvement. We analyse this improvement by examining in detail the effect of the retraining on individual dependency types.