State-of-the-art image classification systems utilize powerful machine-learning-based tools such as Convolutional Neural Networks (CNNs). These networks can achieve high recognition accuracies, but suffer from a black-box problem where the inner workings are incomprehensible by humans that seek to use them. In this paper, a Feature-Oriented Cellular Automata Learning (FOCAL) system is developed to extend traditional gradient-filter-based methods by implementing a Cellular Automata (CA) reasoner utilizing rule-based primitives for determining mutual agreement between neighboring pixels. This novel method is demonstrated to identify features more accurately than standard filter methods and produce classification results that are competitive with typical CNNs, while also allowing a-priori definition of important features facilitating explainable feature classification decision processes. Experiments spanning a variety of influential factors indicate that rebaselining and normalization are vital to the success of the CA-based approach. Furthermore, within certain models, the use of CA is shown to reduce computational demand by over 90% while incurring only a 2% reduction in classification accuracy. Finally, the scalability of the FOCAL system is investigated using the CIFAR-10 dataset and contemporary Deep Neural Networks, and shown to encourage promising avenues of research into explainability while reducing computational processing demands.

Understanding online populations is required for successful human-computer interaction (HCI) research, especially among concerns of replicability among human-subjects research. Scholarly works that rely on online populations must use actionable data to support claims of reliability and bias of populations when providing supposedly objective results. In this work, we provide insight into the tribulations of online subject populations’ ability to complete emotional tasks in particular. We explore the populations of the Amazon Mechanical Turk (MTurk) and Prolific platforms and examine the role of emotion induction in a scenario where subjects play a short online game. We discovered unique differences in the two populations’ self-reports of emotion and dropout rates while other quantitative performance results remained significant across both populations, indicating a more nuanced understanding of the fidelity of these populations than previously reported.

Emotion detection is a rapidly advancing method of quantifying the human experience. Past literature shows emotional data is highly sensitive and private. It also shows habituation effects (previous exposure to a concept) can result in more lenient ethical evaluation of actions. To build effective virtual agents, emotional data must be used in a way that is ethical and inoffensive to humans. Agents which are designed to interact with humans in virtually any capacity should strive to better understand them to be more competitive, more understanding, and generally more effective. We must understand the impact that agents using such data will have on people. To explore these points, we have conducted a 168-participant 2x2 experimental design with an additional user-choice factor to examine effects of pre-exposure to emotion detection on the participant's evaluation of its ethicality. We hypothesize that habituation affects participants' ethical evaluation of the technology presented to them. We found these effects and behavioral impacts when participants played an economic game. We show that the agent's presence on human perception and behavior is significant; proper agent design requires attention to these factors.

Agents that interact with humans in complex, social tasks need the ability to comprehend as well as employ common social strategies. In negotiation, there is ample evidence of such techniques being used efficaciously in human interchanges. In this work, we demonstrate a new design for socially aware agents that employ one such technique---favor exchange---in order to gain value when playing against humans. In an online study of a robust, simulated social negotiation task, we show that these agents are effective against real human participants. In particular, we show that agents that ask for favors during the course of a repeated set of negotiations are more successful than those that do not. Additionally, previous work has demonstrated that humans can detect when agents betray them by failing to return favors that were previously promised. By contrast, this work indicates that these betrayal techniques may go largely undetected in complex scenarios.

We present the results of the 2ndAnnual Human-Agent League of the Automated Negotiating Agent Competition. Building on the success of the previous year's results, a new challenge was issued that focused exploring the likeability-success tradeoff in negotiations. By examining a series of repeated negotiations, actions may affect the relationship between automated negotiating agents and their human competitors over time. The results presented herein support a more complex view of human-agent negotiation and capture of integrative potential (win-win solutions). We show that, although likeability is generally seen as a tradeoff to winning, agents are able to remain well-liked while winning if integrative potential is not discovered in a given negotiation. The results indicate that the top-performing agent in this competition took advantage of this loophole by engaging in favor exchange across negotiations (cross-game logrolling). These exploratory results provide information about the effects of different submitted “black-box” agents in human-agent negotiation and provide a state-of-the-art benchmark for human-agent design.

Recent research shows that how we respond to other social actors depends on what sort of mind we ascribe to them. In this article we examine how perceptions of a virtual agent's mind shape behavior in human-agent negotiations. We varied descriptions and communicative behavior of virtual agents on two dimensions according to the mind perception theory:agency (cognitive aptitude) andpatiency (affective aptitude). Participants then engaged in negotiations with the different agents. People scored more points and engaged in shorter negotiations with agents described to be cognitively intelligent, and got lower points and had longer negotiations with agents that were described to be cognitively unintelligent. Accordingly, agents described as having low agency ended up earning more points than those with high agency. Within the negotiations themselves, participants sent more happy and surprise emojis and emotionally valenced messages to agents described to be emotional. This high degree of described patiency also affected perceptions of the agent's moral standing and relatability. In short, manipulating the perceived mind of agents affects how people negotiate with them. We discuss these results, which show that agents are perceived not only as social actors, but as intentional actors through negotiations.

We present the results of a machine-learning approach to the analysis of several human-agent negotiation studies. By combining expert knowledge of negotiating behavior compiled over a series of empirical studies with neural networks, we show that a hybrid approach to parameter selection yields promise for designing -more effective and socially intelligent agents. Specifically, we show that a deep feedforward neural network using a theory-driven three-parameter model can be effective in predicting negotiation outcomes. Furthermore, it outperforms other expert-designed models that use more parameters, as well as those using other, more limited techniques (such as linear regression models or boosted decision trees). We anticipate these results will have impact for those seeking to combine extensive domain knowledge with more automated approaches in human-computer negotiation.

Research has demonstrated promising benefits of applying virtual trainers to promote physical fitness. The current study investigated the value of virtual agents in the context of personal fitness, compared to trainers with greater levels of perceived agency (avatar or live human). We also explored the possibility that the effectiveness of the virtual trainer might depend on the affective tone it uses when trying to motivate users. Accordingly, participants received either positively or negatively valenced motivational messages from a virtual human they believed to be either an agent or an avatar, or they received the messages from a human instructor via skype. Both self-report and physiological data were collected. Like in-person coaches, the live human trainer who used negatively valenced messages were well-regarded; however, when the agent or avatar used negatively valenced messages, participants responded more poorly than when they used positively valenced ones. Perceived agency also affected rapport: compared to the agent, users felt more rapport with the live human trainer or the avatar. Regardless of trainer type, they also felt more rapport - and said they put in more effort - with trainers that used positively valenced messages than those that used negatively valenced ones. However, in reality, they put in more physical effort (as measured by heart rate) when trainers employed the more negatively valenced affective tone. We discuss implications for human--computer interaction.

We present the results of the first annual Human-Agent League of ANAC. By introducing a new human-agent negotiating platform to the research community at large, we facilitated new advancements in human-aware agents. This has succeeded in pushing the envelope in agent design, and creating a corpus of useful human-agent interaction data. Our results indicate a variety of agents were submitted, and that their varying strategies had distinct outcomes on many measures of the negotiation. These agents approach the problems endemic to human negotiation, including user modeling, bidding strategy, rapport techniques, and strategic bargaining. Some agents employed advanced tactics in information gathering or emotional displays and gained more points than their opponents, while others were considered more "likeable" by their partners.

We present the results of a study in which humans negotiate with computerized agents employing varied tactics over a repeated number of economic ultimatum games. We report that certain agents are highly effective against particular classes of humans: several individual difference measures for the human participant are shown to be critical in determining which agents will be successful. Asking for favors works when playing with pro-social people but backfires with more selfish individuals. Further, making poor offers invites punishment from Machiavellian individuals. These factors may be learned once and applied over repeated negotiations, which means user modeling techniques that can detect these differences accurately will be more successful than those that don't. Our work additionally shows that a significant benefit of cooperation is also present in repeated games---after sufficient interaction. These results have deep significance to agent designers who wish to design agents that are effective in negotiating with a broad swath of real human opponents. Furthermore, it demonstrates the effectiveness of techniques which can reason about negotiation over time.

Virtual agents have been used as tools in negotiation—from acting as mediators to manifesting as full-fledged conversational partners. Virtual agents are a powerful tool for teaching negotiation skills, but require an accurate model of human behavior to perform well both as partners and teachers. The work proposed here aims to expand the current horizon of virtual negotiating agents to utilize human-like strategies. Further agents developed using this framework should be cognizant of the social factors influencing negotiation, including reputation effects and the implications of long-term repeated relationships. A roadmap of current efforts to develop agent platforms and future expansions is discussed.

Negotiation between virtual agents and humans is a complex field that requires designers of systems to be aware not only of the efficient solutions to a given game, but also the mechanisms by which humans create value over multiple negotiations. One way of considering the agent's impact beyond a single negotiation session is by considering the use of external “ledgers” across multiple sessions. We present results that describe the effects of favor exchange on negotiation outcomes, fairness, and trust for two distinct cross-cultural populations, and illustrate the ramifications of their similarities and differences on virtual agent design.