Adaptation Mechanisms in Human–Agent Interaction: Effects on User’s Impressions and Engagement

Adaptation is a key mechanism in human–human interaction. In our work, we aim at endowing embodied conversational agents with the ability to adapt their behavior when interacting with a human interlocutor. With the goal to better understand what the main challenges concerning adaptive agents are, we investigated the effects on the user’s experience of three adaptation models for a virtual agent. The adaptation mechanisms performed by the agent take into account the user’s reaction and learn how to adapt on the fly during the interaction. The agent’s adaptation is realized at several levels (i.e., at the behavioral, conversational, and signal levels) and focuses on improving the user’s experience along different dimensions (i.e., the user’s impressions and engagement). In our first two studies, we aim to learn the agent’s multimodal behaviors and conversational strategies to dynamically optimize the user’s engagement and impressions of the agent, by taking them as input during the learning process. In our third study, our model takes both the user’s and the agent’s past behavior as input and predicts the agent’s next behavior. Our adaptation models have been evaluated through experimental studies sharing the same interacting scenario, with the agent playing the role of a virtual museum guide. These studies showed the impact of the adaptation mechanisms on the user’s experience of the interaction and their perception of the agent. Interacting with an adaptive agent vs. a nonadaptive agent tended to be more positively perceived. Finally, the effects of people’s a priori about virtual agents found in our studies highlight the importance of taking into account the user’s expectancies in human–agent interaction.

Toddlers search longer when there is more information to be gained

One of the greatest challenges for artificial intelligence is how to behave adaptively in scenarios with uncertain or no rewards. One---and perhaps the only---way to approach such complex learning problems is to build simple algorithms that grow into sophisticated adaptive agents, just like children do. But what drives children to explore and learn when external rewards are absent? Across three studies, we tested whether information gain itself acts as an internal reward and motivates children's actions. We measured 24- to 56-month-olds’ persistence in a game where they had to search for an object (animal or toy), which they never find, hidden behind a series of doors, manipulating the degree of uncertainty about \emph{which specific object} was hidden. We found that children were more persistent in their search when there was higher uncertainty, and therefore more information to be gained with each action, highlighting the importance of research on artificial intelligence to invest in curiosity-driven algorithms.

Atomic Dynamic Flow Games: Adaptive vs. Nonadaptive Agents

We propose a game model for selfish routing of atomic agents, who compete for use of a network to travel from their origins to a common destination as quickly as possible. We follow a frequently used rule that the latency an agent experiences on each edge is a constant transit time plus a variable waiting time in a queue. A key feature that differentiates our model from related ones is an edge-based tie-breaking rule for prioritizing agents in queueing when they reach an edge at the same time. We study both nonadaptive agents (each choosing a one-off origin–destination path simultaneously at the very beginning) and adaptive ones (each making an online decision at every nonterminal vertex they reach as to which next edge to take). On the one hand, we constructively prove that a (pure) Nash equilibrium (NE) always exists for nonadaptive agents and show that every NE is weakly Pareto optimal and globally first-in first-out. We present efficient algorithms for finding an NE and best responses of nonadaptive agents. On the other hand, we are among the first to consider adaptive atomic agents, for which we show that a subgame perfect equilibrium (SPE) always exists and that each NE outcome for nonadaptive agents is an SPE outcome for adaptive agents but not vice versa.

Crowd-shipping for urban food rescue logistics

PurposeThe purpose of this research is to gain a better understanding of how a crowd-shipping platform can achieve a critical mass of senders and carrier crowd members to yield network effects that are necessary for the platform to grow and thrive. Specifically, this research studies the participation decisions of both senders and carriers over time and the impacts of the resulting feedback loop on platform growth and performance.Design/methodology/approachAn agent-based model is developed and used to study dynamic behavior and network effects within a simulated crowd-shipping platform. The model allows both carriers and senders to be represented as autonomous, heterogeneous and adaptive agents, whose decisions to participate in the platform impact the participation of other agents over time. Survey data inform the logic governing agent decisions and behaviors.FindingsThe feedback loop created by individual sender and carrier agents' participation decisions generates complex and dynamic network effects that are observable at the platform level. Experimental results demonstrate the importance of having sufficient crowd carriers available when the platform is initially launched, as well as ensuring that sender and carrier participation remains balanced as the platform grows over time.Research limitations/implicationsThe model successfully demonstrates the power of agent-based modeling (ABM) in analyzing network effects in crowd-shipping systems. However, the model has not yet been fully validated with data from a real-world crowd-shipping platform. Furthermore, the model's geographic scope is limited to a single census tract. Platform behavior will likely differ across geographic regions, with varying demographics and sender/carrier density.Practical implicationsThe modeling approach can be used to provide the manager of a volunteer-based crowd-shipping program for food rescue with insights on how to achieve a critical mass of participants, with an appropriate balance between the number of restaurant food donation delivery requests and the number of crowd-shippers available and willing to make those deliveries.Social implicationsThis research can help a crowd-shipping platform for urban food rescue to grow and become self-sustainable, thereby serving more food-insecure people.Originality/valueThe model represents both senders and the carrier crowd as autonomous, heterogeneous and adaptive agents, such that network effects resulting from their interactions can emerge and be observed over time. The model was designed to study a volunteer crowd-shipping platform for food rescue, with participant motivations driven by personal values and social factors, rather than monetary incentives.

Team Synchronization and Individual Contributions in Coop-Space Fortress

This work studied human teamwork with a concentration on the influence of team synchronization and in- dividual differences on performance. Human participants were paired to complete collaborative tasks in a simulated game environment, in which they were assigned roles with corresponding responsibilities. Cross- correlation analysis was employed to quantify the degree of team synchronization and time-lag between two teammates’ collective actions. Results showed that team performance is determined by factors at both the individual and team levels. We found interaction effects between team synchronization and individual differences and quantified their contributions to team performance. The application of our research findings and proposed quantitative methods for developing adaptive agents for human-autonomy teaming is discussed.

Behavior Morphologies of Machine Learning Agents in Media Artworks

Since the 1950s, a range of artists have used artificial agents in their work, in parallel with scientific research in cybernetics, artificial intelligence (AI) and artificial life (AL). In particular, an increasing number of artists work with machine learning and other adaptive systems. Through my own engagement with such systems, I analyze adaptive agents within the broader context of the aesthetic of behavior. As a result, I propose an aesthetic framework for understanding behaviors which considers the role of the observer as an adaptive perceiving agent, the unfathomable character of machine learning systems, and the morphology of behaviors as time-based phenomenon.