This paper presents preliminary results of our work with a major financial company, where we try to use methods of plan recognition in order to investigate the interactions of a costumer with the company's online interface. In this paper, we present the first steps of integrating a plan recognition algorithm in a real-world application for detecting and analyzing the interactions of a costumer. It uses a novel approach for plan recognition from bare-bone UI data, which reasons about the plan library at the lowest recognition level in order to define the relevancy of actions in our domain, and then uses it to perform plan recognition. We present preliminary results of inference on three different use-cases modeled by domain experts from the company, and show that this approach manages to decrease the overload of information required from an analyst to evaluate a costumer's session - whether this is a malicious or benign session, whether the intended tasks were completed, and if not - what actions are expected next.

A key challenge in procedural content generation is to automatically evaluate whether the generated content has good quality. In this paper, we describe an approach that uses nonexpert workers to evaluate small portions of levels generated by an off-the-shelf generation system for the game of Infinite Mario Bros. Several such evaluated portions are then combined to form full levels of the game using a mathematical progression arc model. The composition of the small portions into full levels is done by accounting for the human-annotated information. We evaluated the approach using computational metrics as well as surveying human subjects playing the levels. The results show that the human computation approach is able to generate levels that are perceived by people to have better visual aesthetics and to be more enjoyable to play than existing approaches. Another contribution of our paper is a dataset of the small annotated levels that can be used in future research for learning models for evaluating machine-generated content.

In many networked decision-making settings, information about the world is distributed across multiple agents and agents’ success depends on their ability to aggregate and reason about their local information over time. This paper presents a computational model of information aggregation in such settings in which agents’ utilities depend on an unknown event. Agents initially receive a noisy signal about the event and take actions repeatedly while observing the actions of their neighbors in the network at each round. Such settings characterize many distributed systems such as sensor networks for intrusion detection and routing systems for Internet traffic. Using the model, we show that (1) agents converge in action and in knowledge for a general class of decision-making rules and for all network structures; (2) all networks converge to playing the same action regardless of the network structure; and (3) for particular network configurations, agents can converge to the correct action when using a well-defined class of myopic decision rules. These theoretical results are also supported by a new simulation-based open-source empirical test-bed for facilitating the study of information aggregation in general networks.

This paper addresses the problem of automated advice provision in scenarios that involve repeated interactions between people and computer agents. This problem arises in many applications such as route selection systems, office assistants and climate control systems. To succeed in such settings agents must reason about how their advice influences people’s future actions or decisions over time. This work models such scenarios as a family of repeated bilateral interaction called “choice selection processes”, in which humans or computer agents may share certain goals, but are essentially self-interested. We propose a social agent for advice provision (SAP) for such environments that generates advice using a social utility function which weighs the sum of the individual utilities of both agent and human participants. The SAP agent models human choice selection using hyperbolic discounting and samples the model to infer the best weights for its social utility function. We demonstrate the effectiveness of SAP in two separate domains which vary in the complexity of modeling human behavior as well as the information that is available to people when they need to decide whether to accept the agent’s advice. In both of these domains, we evaluated SAP in extensive empirical studies involving hundreds of human subjects. SAP was compared to agents using alternative models of choice selection processes informed by behavioral economics and psychological models of decision-making. Our results show that in both domains, the SAP agent was able to outperform alternative models. This work demonstrates the efficacy of combining computational methods with behavioral economics to model how people reason about machine-generated advice and presents a general methodology for agent-design in such repeated advice settings.

Despite the abundance of strategies in the multi-agent systems literature on repeated negotiation under incomplete information, there is no single negotiation strategy that is optimal for all possible domains. Thus, agent designers face an “algorithm selection” problem—which negotiation strategy to choose when facing a new domain and unknown opponent. Our approach to this problem is to design a “meta-agent” that predicts the performance of different negotiation strategies at run-time. We study two types of the algorithm selection problem in negotiation: In the off-line variant, an agent needs to select a negotiation strategy for a given domain but cannot switch to a different strategy once the negotiation has begun. For this case, we use supervised learning to select a negotiation strategy for a new domain that is based on predicting its performance using structural features of the domain. In the on-line variant, an agent is allowed to adapt its negotiation strategy over time. For this case, we used multi-armed bandit techniques that balance the exploration–exploitation tradeoff of different negotiation strategies. Our approach was evaluated using the GENIUS negotiation test-bed that is used for the annual international Automated Negotiation Agent Competition which represents the chief venue for evaluating the state-of-the-art multi-agent negotiation strategies. We ran extensive simulations using the test bed with all of the top-contenders from both off-line and on-line negotiation tracks of the competition. The results show that the meta-agent was able to outperform all of the finalists that were submitted to the most recent competition, and to choose the best possible agent (in retrospect) for more settings than any of the other finalists. This result was consistent for both off-line and on-line variants of the algorithm selection problem. This work has important insights for multi-agent systems designers, demonstrating that “a little learning goes a long way”, despite the inherent uncertainty associated with negotiation under incomplete information.

Choice selection processes are a family of bilateral games of incomplete information in which a computer agent generates advice for a human user while considering the effect of the advice on the user's behavior in future interactions. The human and the agent may share certain goals, but are essentially self-interested. This paper extends selection processes to settings in which the actions available to the human are ordered and thus the user may be influenced by the advice even though he doesn't necessarily follow it exactly. In this work we also consider the case in which the user obtains some observation on the sate of the world. We propose several approaches to model human decision making in such settings. We incorporate these models into two optimization techniques for the agent advice provision strategy. In the first one the agent used a social utility approach which considered the benefits and costs for both agent and person when making suggestions. In the second approach we simplified the human model in order to allow modeling and solving the agent strategy as an MDP. In an empirical evaluation involving human users on AMT, we showed that the social utility approach significantly outperformed the MDP approach.

The Simple-Meta agent uses machine learning to select the negotiation strategy that is predicted to be most successful based on structural features of the domain.

Despite committees and elections being widespread in thereal-world, the design of agents for operating in humancomputer committees has received far less attention than thetheoretical analysis of voting strategies. We address this gapby providing an agent design that outperforms other voters ingroups comprising both people and computer agents. In oursetting participants vote by simultaneously submitting a ranking over a set of candidates and the election system uses a social welfare rule to select a ranking that minimizes disagreements with participants’ votes. We ran an extensive studyin which hundreds of people participated in repeated votingrounds with other people as well as computer agents that differed in how they employ strategic reasoning in their votingbehavior. Our results show that over time, people learn todeviate from truthful voting strategies, and use heuristics toguide their play, such as repeating their vote from the previous round. We show that a computer agent using a bestresponse voting strategy was able to outperform people in thegame. Our study has implication for agent designers, highlighting the types of strategies that enable agents to succeedin committees comprising both human and computer participants. This is the first work to study the role of computeragents in voting settings involving both human and agent participants.

In exploratory domains, agents' actions map onto logs of behavior that include switching between activities, extraneous actions, and mistakes. These aspects create a challenging plan recognition problem. This paper presents a new algorithm for inferring students' activities in exploratory domains that is evaluated empirically using a new type of flexible and open-ended educational software for science education. Such software has been shown to provide a rich educational environment for students, but challenge teachers to keep track of students' progress and to assess their performance. The algorithm decomposes students complete interaction histories to create hierarchies of interdependent tasks that describe their activities using the software. It matches students' actions to a predefined grammar in a way that reflects that students solve problems in a modular fashion but may still interleave between their activities. The algorithm was empirically evaluated on peoples interaction with two separate software systems for simulating a chemistry laboratory and for statistics education. It was separately compared to the state-of-the-art recognition algorithms for each of the software. The results show that the algorithm was able to correctly infer students' activities significantly more often than the state-of-the-art, and was able to generalize to both of the software systems with no intervention.

This article studies how people reveal private information in strategic settings in which participants need to negotiate over resources but are uncertain about each other's objectives. The study compares two negotiation protocols that differ in whether they allow participants to disclose their objectives in a repeated negotiation setting of incomplete information. Results show that most people agree to reveal their goals when asked, and this leads participants to more beneficial agreements. Machine learning was used to model the likelihood that people reveal their goals in negotiation, and this model was used to make goal request decisions in the game. In simulation, use of this model is shown to outperform people making the same type of decisions. These results demonstrate the benefit of this approach towards designing agents to negotiate with people under incomplete information.

The rapid dissemination of technology such as the Internet across geographical and ethnic lines is opening up opportunities for computer agents to negotiate with people of diverse cultural and organizational affiliations. To negotiate proficiently with people in different cultures, agents need to be able to adapt to the way behavioral traits of other participants change over time. This article describes a new agent for repeated bilateral negotiation that was designed to model and adapt its behavior to the individual traits exhibited by its negotiation partner. The agent’s decision-making model combined a social utility function that represented the behavioral traits of the other participant, as well as a rule-based mechanism that used the utility function to make decisions in the negotiation process. The agent was deployed in a strategic setting in which both participants needed to complete their individual tasks by reaching agreements and exchanging resources, the number of negotiation rounds was not fixed in advance and agreements were not binding. The agent negotiated with human subjects in the United States and Lebanon in situations that varied the dependency relationships between participants at the onset of negotiation. There was no prior data available about the way people would respond to different negotiation strategies in these two countries. Results showed that the agent was able to adopt a different negotiation strategy to each country. Its average performance across both countries was equal to that of people. However, the agent outperformed people in the United States, because it learned to make offers that were likely to be accepted by people, while being more beneficial to the agent than to people. In contrast, the agent was outperformed by people in Lebanon, because it adopted a high reliability measure which allowed people to take advantage of it. These results provide insight for human-computer agent designers in the types of multicultural settings that we considered, showing that adaptation is a viable approach towards the design of computer agents to negotiate with people when there is no prior data of their behavior.

This paper describes a challenging plan recognition problem that arises in environments in which agents engage widely in exploratory behavior, and presents new algorithms for effective plan recognition in such settings. In exploratory domains, agentsʼ actions map onto logs of behavior that include switching between activities, extraneous actions, and mistakes. Flexible pedagogical software, such as the application considered in this paper for statistics education, is a paradigmatic example of such domains, but many other settings exhibit similar characteristics. The paper establishes the task of plan recognition in exploratory domains to be NP-hard and compares several approaches for recognizing plans in these domains, including new heuristic methods that vary the extent to which they employ backtracking, as well as a reduction to constraint-satisfaction problems. The algorithms were empirically evaluated on peopleʼs interaction with flexible, open-ended statistics education software used in schools. Data was collected from adults using the software in a lab setting as well as middle school students using the software in the classroom. The constraint satisfaction approaches were complete, but were an order of magnitude slower than the heuristic approaches. In addition, the heuristic approaches were able to perform within 4% of the constraint satisfaction approaches on student data from the classroom, which reflects the intended user population of the software. These results demonstrate that the heuristic approaches offer a good balance between performance and computation time when recognizing peopleʼs activities in the pedagogical domain of interest.

The ability to negotiate successfully is critical in many social interactions. The dissemination of applications such as the Internet across geographical and ethnic borders are opening up opportunities for computer agents to negotiate with people of diverse cultural and organizational affiliation. These automated negotiators should be able to proficiently interact and collaborate with their human partners. In this paper we compare several techniques for modeling the negotiation behavior of people across three different countries. Culture plays an important role in people’s decision making and people differ in the way make offers and fulfil their commitments in negotiation across cultures. We consider a setting that included multiple rounds of negotiation with non-binding agreements. Participants in each of the countries interacted with a computer agent that used a baseline negotiation strategy that adapted to the extent to which participants were helpful and reliable. The models considered various features of the negotiation task, such as the extent to which proposals are generous and helpful to participants and whether participants fulfil their agreements. Our models achieved high accuracy rates in predicting important decisions that are made in negotiation such as whether participants reach agreement and the extent to which they are reliable in different situations. We show that the features that best vary the prediction accuracy depend on people’s cultural affiliation in addition to their actual negotiation behavior. These models for predicting human behavior in negotiation will form the basis of a computer agent that can successfully negotiate with people across cultures.

Computer systems increasingly carry out tasks in mixed networks, that is in group settings in which they interact both with other computer systems and with people. Participants in these heterogeneous human–computer groups vary in their capabilities, goals, and strategies; they may cooperate, collaborate, or compete. The presence of people in mixed networks raises challenges for the design and the evaluation of decision-making strategies for computer agents. This paper describes several new decision-making models that represent, learn and adapt to various social attributes that influence people's decision-making and presents a novel approach to evaluating such models. It identifies a range of social attributes in an open-network setting that influence people's decision-making and thus affect the performance of computer-agent strategies, and establishes the importance of learning and adaptation to the success of such strategies. The settings vary in the capabilities, goals, and strategies that people bring into their interactions. The studies deploy a configurable system called Colored Trails (CT) that generates a family of games. CT is an abstract, conceptually simple but highly versatile game in which players negotiate and exchange resources to enable them to achieve their individual or group goals. It provides a realistic analogue to multi-agent task domains, while not requiring extensive domain modeling. It is less abstract than payoff matrices, and people exhibit less strategic and more helpful behavior in CT than in the identical payoff matrix decision-making context. By not requiring extensive domain modeling, CT enables agent researchers to focus their attention on strategy design, and it provides an environment in which the influence of social factors can be better isolated and studied.

Ideally designed software allow users to explore and pursue interleaving plans, making it challenging to automatically recognize user interactions. The recognition algorithms presented use constraint satisfaction techniques to compare user interaction histories to a set of ideal solutions. We evaluate these algorithms on data obtained from user interactions with a commercially available pedagogical software, and find that these algorithms identified users’ activities with 93% accuracy.

Natural Intelligence is based not only on conscious procedural anddeclarative knowledge, but also on knowledge that is inferred fromobserving the actions of others. This knowledge is tacit, in that theprocess of its acquisition remains unspecified. However, tacitknowledge is an accepted guide of behavior, especiallyin unfamiliar contexts. In situations where knowledge is lacking,animals act on these beliefs without explicitly reasoning about theworld or fully considering the consequences of their actions. Thispaper provides a computational model of behavior in which tacitknowledge plays a crucial role. We model how knowledge arises fromobserving different types of agents, each of whom reacts differentlyto the behaviors of others in an unfamiliar context. Agents'interaction in this context is described using directed graphs. Weshow how a set of observations guide agents' knowledge and behaviorgiven different states of the world.

This paper presents a model of collaborative decision-making for groups that involve people and computer agents. The model distinguishes between actions relating to participants’ commitment to the group and actions relating to their individual tasks, uses this distinction to decompose group decision making into smaller problems that can be solved efficiently. It allows computer agents to reason about the benefits of their actions on a collaboration and the ways in which human participants perceive these benefits. The model was tested in a setting in which computer agents need to decide whether to interrupt people to obtain potentially valuable information. Results show that the magnitude of the benefit of interruption to the collaboration is a major factor influencing the likelihood that people will accept interruption requests. They further establish that people’s perceived type of their partners (whether humans or computers) significantly affected their perceptions of the usefulness of interruptions when the benefit of the interruption is not clear-cut. These results imply that system designers need to consider not only the possible benefits of interruptions to collaborative human-computer teams but also the way that such benefits are perceived by people.

This paper considers the design of agent strategies for deciding whether to help other members of a group with whom an agent is engaged in a collaborative activity. Three characteristics of collaborative planning must be addressed by these decision-making strategies: agents may have only partial information about their partners' plans for sub-tasks of the collaborative activity; the effectiveness of helping may not be known a priori; and, helping actions have some associated cost. The paper proposes a novel probabilistic representation of other agents' beliefs about the recipes selected for their own or for the group activity, given partial information. This representation is compact, and thus makes reasoning about helpful behavior tractable. The paper presents a decision-theoretic mechanism that uses this representation to make decisions about two kinds of helpful actions: communicating information relevant to a partner's plans for some sub-action, and adding domain actions that are helpful to other agent(s) into the collaborative plan. This mechanism includes a set of rules for reasoning about the utility of helpful actions and the cost incurred by doing them. It was tested using a multi-agent test-bed with configurations that varied agents' uncertainty about the world, their uncertainty about each others' capabilities or resources, and the cost of helpful behavior. In all cases, agents using the decision-theoretic mechanism to decide whether to help outperformed agents using purely axiomatic rules.

This paper addresses the problem of learning with whom to interact in situations where obtaining information about others is associated with a cost, and this information is potentially unreliable. It considers settings in which agents need to decide whether to engage in a series of interactions with partners of unknown competencies, and can purchase reports about partners' competencies from others. The paper shows that Hierarchical Bayesian models offer a unified approach for (1) inferring the reliability of information providers, and (2) learning the competencies of individual agents as well as the general population. The performance of this model was tested in experiments of varying complexity, measuring agents' performance as well as error in estimating others' competencies. Results show that agents using the hierarchical model to make decisions outperformed other probabilistic models from the recent literature, even when there was a high ratio of unreliable information providers.