Human Modeling and Prediction

Human collaborator’s workload plays a central role in human-robot collaboration. Algorithms designed to minimize cognitive workload enhance fluent human-robot teamwork. Time series data of workload is vital for both designing and assessing these algorithms. However, accurately quantifying and measuring cognitive workload, particularly at high temporal resolution, poses a substantial challenge. Towards addressing this challenge, we explore the potential of after-action reviews (AARs) as a tool for gauging workload during human-robot collaboration. First, through a case study, we present and demonstrate AutoAAR for measuring human workload post-task at a high temporal resolution. Second, through a user study, we quantify the validity and utility of measurements derived using AutoAAR for human-robot teamwork.The paper concludes with guidelines and future directions to extend this method to measure other internal states, such as trust and intent.

When faced with accomplishing a task, human experts exhibit intentional behavior. Their unique intents shape their plans and decisions, resulting in experts demonstrating diverse behaviors to accomplish the same task. Due to the uncertainties encountered in the real world and their bounded rationality, experts sometimes adjust their intents, which in turn influences their behaviors during task execution. This paper introduces IDIL, a novel imitation learning algorithm to mimic these diverse intent-driven behaviors of experts. Iteratively, our approach estimates expert intent from heterogeneous demonstrations and then uses it to learn an intent-aware model of their behavior. Unlike contemporary approaches, IDIL is capable of addressing sequential tasks with high-dimensional state representations, while sidestepping the complexities and drawbacks associated with adversarial training (a mainstay of related techniques). Our empirical results suggest that the models generated by IDIL either match or surpass those produced by recent imitation learning benchmarks in metrics of task performance. Moreover, as it creates a generative model, IDIL demonstrates superior performance in intent inference metrics, crucial for human-agent interactions, and aptly captures a broad spectrum of expert behaviors.

To forge effective collaborations with humans, robots require the capacity to understand and predict the behaviors of their human counterparts. There is a growing body of computational research on human modeling for human-robot interaction (HRI). However, a key bottleneck in conducting this research is the relative lack of data of human internal states – like intent, workload, and trust – which undeniably affect human behavior. Despite their significance, these states are elusive to measure, making the assembly of datasets a challenge and hindering the progression of human modeling techniques. To help address this, we first introduce Rescue World for Teams (RW4T): a configurable testbed to simulate disaster response scenarios requiring human-robot collaboration. Next, using RW4T, we curate a multimodal dataset of human-robot behavior and internal states in dyadic human-robot collaboration. This RW4T dataset includes state, action and reward sequences, and all the necessary data to replay a visual task execution. It further contains psychophysiological metrics like heart rate and pupillometry, complemented by self-reported cognitive state measures. With data from 20 participants, each undertaking five human-robot collaborative tasks, this dataset accompanied with the simulator can serve as a valuable benchmark for human behavior modeling.

We present Bayesian Team Imitation Learner (BTIL), an imitation learning algorithm to model the behavior of teams performing sequential tasks in Markovian domains. In contrast to existing multi-agent imitation learning techniques, BTIL explicitly models and infers the time-varying mental states of team members, thereby enabling learning of decentralized team policies from demonstrations of suboptimal teamwork. Further, to allow for sample- and label-efficient policy learning from small datasets, BTIL employs a Bayesian perspective and is capable of learning from semi-supervised demonstrations. We demonstrate and benchmark the performance of BTIL on synthetic multi-agent tasks as well as a novel dataset of human-agent teamwork. Our experiments show that BTIL can successfully learn team policies from demonstrations despite the influence of team members’ (time-varying and potentially misaligned) mental states on their behavior.

Autonomous agents operating in the real world often need to interact with other agents to accomplish their tasks. For such agents, the ability to model behavior of other agents – both human and artificial – without complete knowledge of their decision factors is essential. Towards realizing this ability, we present Factorial Agent Markov Model (FAMM), a model to represent behavior of other agents performing sequential tasks. In contrast with most existing models, FAMM allows for behavior of other agents to depend on multiple, time-varying latent decision factors and does not assume rationality. To enable learning of \famm parameters by observing behavior of other agents, we provide a set of variational inference algorithms for the unsupervised, semi-supervised, and supervised settings. These Bayesian learning algorithms for the FAMM enable agents to model other agents using execution traces and domain-specific priors. We demonstrate the utility of FAMM and corresponding learning algorithms using three synthetic domains and benchmark them against existing algorithms for modeling agent behavior. Our numerical experiments demonstrate that, despite the presence of multiple and time-varying latent states, our approach is capable of learning predictive models of other agents with semi-supervision.

Trained AI systems and expert decision makers can make errors that are often difficult to identify and understand. Determining the root cause for these errors can improve future decisions. This work presents Generative Error Model (GEM), a generative model for inferring representational errors based on observations of an actor’s behavior (either simulated agent, robot, or human). The model considers two sources of error: those that occur due to representational limitations – "blind spots" – and non-representational errors, such as those caused by noise in execution or systematic errors present in the actor’s policy. Disambiguating these two error types allows for targeted refinement of the actor’s policy (i.e., representational errors require perceptual augmentation, while other errors can be reduced through methods such as improved training or attention support). We present a Bayesian inference algorithm for GEM and evaluate its utility in recovering representational errors on multiple domains. Results show that our approach can recover blind spots of both reinforcement learning agents as well as human users.

Artificial agents that interact with other (human or artificial) agents require models in order to reason about those other agents’ behavior. In addition to the predictive utility of these models, maintaining a model that is aligned with an agent’s true generative model of behavior is critical for effective human-agent interaction. In applications wherein observations and partial specification of the agent’s behavior are available, achieving model alignment is challenging for a variety of reasons. For one, the agent’s decision factors are often not completely known; further, prior approaches that rely upon observations of agents’ behavior alone can fail to recover the true model, since multiple models can explain observed behavior equally well. To achieve better model alignment, we provide a novel approach capable of learning aligned models that conform to partial knowledge of the agent’s behavior. Central to our approach are a factored model of behavior (AMM), along with Bayesian nonparametric priors, and an inference approach capable of incorporating partial specifications as constraints for model learning. We evaluate our approach in experiments and demonstrate improvements in metrics of model alignment.