Stackelberg games have been used in several deployed applications of game theory to make recommendations for allocating limited resources for protecting critical infrastructure. The resource allocation strategies are randomized to prevent a strategic attacker from using surveillance to learn and exploit patterns in the allocation. An important limitation of previous work on security games is that it typically assumes that attackers have perfect surveillance capabilities, and can learn the exact strategy of the defender. We introduce a new model that explicitly models the process of an attacker observing a sequence of resource allocation decisions and updating his beliefs about the defender's strategy. For this model we present computational techniques for updating the attacker's beliefs and computing optimal strategies for both the attacker and defender, given a specific number of observations. We provide multiple formulations for computing the defender's optimal strategy, including non-convex programming and a convex approximation. We also present an approximate method for computing the optimal length of time for the attacker to observe the defender's strategy before attacking. Finally, we present experimental results comparing the efficiency and runtime of our methods.

This paper extends hierarchical task network (HTN) planning with lightweight learning, considering that in robotics, actions have a non-zero probability of failing. Our work applies to A*-based HTN planners with lifting. We prove that the planner finds the plan of maximal expected utility, while retaining its lifting capability and efficient heuristic-based search. We show how to learn the probabilities online, which allows a robot to adapt by replanning on execution failures. The idea behind this work is to use the HTN domain to constrain the space of possibilities, and then to learn on the constrained space in a way requiring few training samples, rendering the method applicable to autonomous mobile robots.

This paper surveys research in robotics in the AAMAS (Au- tonomous Agents and Multi-Agent Systems) community. It argues that the autonomous agents community can, and has, impact on robotics. Moreover, it argues that agents re- searchers should proactively seek to impact the robotics com- munity, to prevent independent re-discovery of known results, and to benefit autonomous agents science. To support these claims, I provide evidence from my own research into multi- robot teams, and from others’.

The paper discusses how a human-robot team can remain “cohesive” while performing under stress. By cohesive the paper understands the ability of the team to operate effectively, with individual members being interdependent-yet-autonomous in carrying out tasks. For a human-robot team, we argue that this requires robots to (1) have an adequate sense of that interde- pendency in terms of the social dynamics within the team, and to (2) maintain transparency towards the human team members in terms of what it is doing, why, and to what extent it can achieve its (possibly jointly agreed upon) goals. The paper re- ports of recent field experience showing that failure in trans- parency results in reduced acceptability of robot autonomous behavior by the human team members. This reduction in acceptability can have two negative impacts on cohesiveness: Humans and robots fail to maintain common ground, and as a result they fail to maintain trust.

Hybrid group autonomy, organizations and teams composed of humans, machines and robots, are important to AI. Unlike the war in Iraq in 2002, the war in Afghanistan has hundreds of mobile robots aloft, on land, or under the sea. But when it comes to solving problems as part of a team, these agents are socially passive. Were the problem of aggregation and the autonomy of hybrids to be solved, robot teams could accompa- ny humans to address and solve problems together on Mars, under the sea, or in dan- gerous locations on earth (such as, fire-fighting, reactor meltdowns, and future wars). “Robot autonomy is required because one soldier cannot control several robots ... [and] because no computational system can discriminate between combatants and innocents in a close-contact encounter.” (Sharkey, 2008) Yet, today, one of the fundamental unsolved problems in the social sciences is the aggregation of individual data (such as preferences) into group (team) data (Giles, 2011) The original motivation behind game theory was to study the effect that multi- ple agents have on each other (Von Neumann and Morgenstern, 1953), known as interdependence or mutual dependence. Essentially, the challenge addresses the ques- tion: why is a group different from the collection of individuals who comprise the group? That the problem remains unsolved almost 70 years later is a remarkable com- ment on the state of the social sciences today, including game theory and economics. But solving this challenge is essential for the science and engineering of multiagent, multirobot and hybrid environments (that is, humans, machines and robots working together).

One of the original applications of crowdsourcing the construction of knowledge is Wikipedia, which relies entirely on people to contribute, extend, and modify the representation of knowledge. This paper presents a case for combining AI and wisdom of the crowds for the social construction of knowledge. Our social-computational approach to collective intelligence combines the strengths of human cognitive diversity in producing content and the capabilities of an AI, through methods such as topic modeling, to link and synthesize across these human contributions. In addition to drawing from established domains such as Wikipedia for inspiration and guidance, we present the design of a system that incorporates AI into wisdom of the crowds to develop a knowledge base on sustainability. In this setting the AI plays the role of scholar, as might many of the other participants, drawing connections and synthesizing across contributions. We close with a general discussion, speculating on educational implications and other roles that an AI can play within an otherwise collective human intelligence.

Crowdsourcing requires new strategies to evaluate the workers involved as well as the quality of workers’ output. Using customer feedback data, we introduce multi-facetted Rasch scaling as an evaluation technique to assess the contributions of workers and products simultaneously within a single coherent measurement framework. Based on a data set of about 250,000 customers who rated nearly 115,000 products, for a total of nearly 3 million cases, we found that product ratings reflect almost as much the existence of stable rater differences as they are indicative of the products’ properties. We illustrate how Rasch scaling provides extensive quality control mechanisms; as well we show how aberrant workers and products can be identified so that appropriate feedback and/or corrective actions can be initiated.

This paper presents work using crowdsourcing to assess explanations for supervised text classification. In this paper, an explanation is defined to be a set of words from the input text that a classifier or human believes to be most useful for making a classification decision. We compared two types of explanations for classification decisions: human-generated and computer-generated. The comparison is based on whether the type of the explanation was identifiable and on which type of explanation was preferred. Crowdsourcing was used to collect two types of data for these experiments. First, human-generated explanations were collected by having users select an appropriate category for a piece of text and highlight words that best support this category. Second, users were asked to compare human- and computer-generated explanations and indicate which they preferred and why. The crowdsourced data used for this paper was collected primarily via Amazon’s Mechanical Turk, using several quality control methods. We found that in one test corpus, the two explanation types were virtually indistinguishable, and that participants did not have a significant preference for one type over another. For another corpus, the explanations were slightly more distinguishable, and participants preferred the computer-generated explanations at a small, but statistically significant, level. We conclude that computer-generated explanations for text classification can be comparable in quality to human-generated explanations.

The current era of internet-facilitated bigger data, better tools, and collective intelligence community computing is accelerating advances in many areas ranging from artificial intelligence to knowledge generation to public health. In the health sector, data volumes are growing with genomic, phenotypic, microbiomic, metabolomic, self-tracking, and other data streams. Simultaneously, tools are proliferating to allow individuals and groups to make sense of these data in a participatory manner through personal health tracking devices, mobile health applications, and personal electronic medical records. Health community computing models are emerging to support individual activity and mass collaboration through health social networks and crowdsourced health research studies. Participatory health efforts portend important benefits based on both size and speed. Studies can be carried out in cohorts of thousands instead of hundreds, and it could be possible to apply findings from newly-published studies with near-immediate speed. One operator of interventional crowdsourced health research studies, DIYgenomics, has several crowdsourced health research studies in open enrollment as of January 2012 in the areas of vitamin deficiency, aging, mental performance, and epistemology. The farther future of intelligent health community computing could include personal health dashboards, continuous personal health information climates, personal virtual coaches (e.g.; Siri 2.0), and an efficient health frontier of dynamic personalized health recommendations and action-taking.

This paper focuses on age-related change in sleep and improves our sleep estimation method by employing the feature of such relation between sleep stage and age. In particular, the wake stage increases as the age increases, while Non-REM stage decrease as the age increase. Using such distinctive features, we propose a new determination sleep stages, and introduce it into for our sleep estimation method based on Genetic Algorithms (GAs), which evolve the sleep stage for each person according to the fitness. To investigate an effectiveness of a new determination of sleep stages, we compare the estimated sleep stages of our method employing the proposed fitness function with that of Hirose’s method. The experimental results suggest that our method employed the proposed discretization of sleep stages has a capability to estimate the sleep stage accurately than Hirose’s method.