We anticipate increased instances of humans and AI systems working together in what we refer to as a hybrid team. The increase in collaboration is expected as AI systems gain proficiency and their adoption becomes more widespread. However, their behavior is not error-free, making hybrid teams a very suitable solution. As such, we consider methods for improving performance for these teams of humans and AI systems. For hybrid teams, we will refer to both the humans and AI systems as agents. To improve team performance over that seen for agents operating individually, we propose a manager which learns, through a standard Reinforcement Learning scheme, how to best delegate, over time, the responsibility of taking a decision to any of the agents. We further guide the manager's learning so they also minimize how many changes in delegation are made resulting from undesirable team behavior. We demonstrate the optimality of our manager's performance in several grid environments which include failure states which terminate an episode and should be avoided. We perform our experiments with teams of agents with varying degrees of acceptable risk, in the form of proximity to a failure state, and measure the manager's ability to make effective delegation decisions with respect to its own risk-based constraints, then compare these to the optimal decisions. Our results show our manager can successfully learn desirable delegations which result in team paths near/exactly optimal with respect to path length and number of delegations.

When humans and autonomous systems operate together as what we refer to as a hybrid team, we of course wish to ensure the team operates successfully and effectively. We refer to team members as agents. In our proposed framework, we address the case of hybrid teams in which, at any time, only one team member (the control agent) is authorized to act as control for the team. To determine the best selection of a control agent, we propose the addition of an AI manager (via Reinforcement Learning) which learns as an outside observer of the team. The manager learns a model of behavior linking observations of agent performance and the environment/world the team is operating in, and from these observations makes the most desirable selection of a control agent. From our review of current state of the art, we present a novel manager model for oversight of hybrid teams by our support for diverse agents and decision-maker operations across multiple time steps and decisions. In our model, we restrict the manager’s task by introducing a set of constraints. The manager constraints indicate acceptable team operation, so a violation occurs if the team enters a condition which is unacceptable and requires manager intervention. To ensure minimal added complexity or potential inefficiency for the team, the manager should attempt to minimize the number of times the team reaches a constraint violation and requires subsequent manager intervention. Therefore, our manager is optimizing its selection of authorized agents to boost overall team performance while minimizing the frequency of manager intervention. We demonstrate our manager’s performance in a simulated driving scenario representing the case of a hybrid team of agents composed of a human driver and autonomous driving system. We perform experiments for our driving scenario with interfering vehicles, indicating the need for collision avoidance and proper speed control. Our results indicate a positive impact on our manager, with some cases resulting in increased team performance up to \(\approx 187\%\) that of the best solo agent performance.

Given the popular presupposition of human reasoning as the standard for learning and decision making, there have been significant efforts and a growing trend in research to replicate these innate human abilities in artificial systems. As such, topics including Game Theory, Theory of Mind, and Machine Learning, among others, integrate concepts that are assumed components of human reasoning. These serve as techniques to replicate and understand the behaviors of humans. In addition, next-generation autonomous and adaptive systems will largely include AI agents and humans working together as teams. To make this possible, autonomous agents will require the ability to embed practical models of human behavior, allowing them not only to replicate human models as a technique to “learn” but also to understand the actions of users and anticipate their behavior, so as to truly operate in symbiosis with them. The main objective of this article is to provide a succinct yet systematic review of important approaches in two areas dealing with quantitative models of human behaviors. Specifically, we focus on (i) techniques that learn a model or policy of behavior through exploration and feedback, such as Reinforcement Learning, and (ii) directly model mechanisms of human reasoning, such as beliefs and bias, without necessarily learning via trial and error.

With humans interacting with AI-based systems at an increasing rate, it is necessary to ensure the artificial systems are acting in a manner which reflects understanding of the human. In the case of humans and artificial AI agents operating in the same environment, we note the significance of comprehension and response to the actions or capabilities of a human from an agent's perspective, as well as the possibility to delegate decisions either to humans or to agents, depending on who is deemed more suitable for a given context. Such capabilities will ensure an improved responsiveness and utility of the entire human-AI system. To that end, we investigate the use of cognitively inspired models of behavior to predict the behavior of both human and AI agents. The predicted behavior, and associated performance with respect to a certain goal, is used to delegate control between humans and AI agents through the use of an intermediary entity. As we demonstrate, this allows overcoming potential shortcomings of either humans or agents in the pursuit of a goal.

With humans interacting with AI-based systems at an increasing rate, it is necessary to ensure the artificial systems are acting in a manner which reflects understanding of the human. In the case of humans and artificial AI agents operating in the same environment, we note the significance of comprehension and response to the actions or capabilities of a human from an agent's perspective, as well as the possibility to delegate decisions either to humans or to agents, depending on who is deemed more suitable at a certain point in time. Such capabilities will ensure an improved responsiveness and utility of the entire human-AI system. To that end, we investigate the use of cognitively inspired models of behavior to predict the behavior of both human and AI agents. The predicted behavior, and associated performance with respect to a certain goal, is used to delegate control between humans and AI agents through the use of an intermediary entity. As we demonstrate, this allows overcoming potential shortcomings of either humans or agents in the pursuit of a goal.

We present Rushmore, a system that securely displays static or animated images using TrustZone. The core functionality of Rushmore is to securely decrypt and display encrypted images (sent by a trusted party) on a mobile device. Although previous approaches have shown that it is possible to securely display encrypted images using TrustZone, they exhibit a critical limitation that significantly hampers the applicability of using TrustZone for display security. The limitation is that, when the trusted domain of TrustZone (the secure world) takes control of the display, the untrusted domain (the normal world) cannot display anything simultaneously. This limitation comes from the fact that previous approaches give the secure world exclusive access to the display hardware to preserve security. With Rushmore, we overcome this limitation by leveraging a well-known, yet overlooked hardware feature called an IPU (Image Processing Unit) that provides multiple display channels. By partitioning these channels across the normal world and the secure world, we enable the two worlds to simultaneously display pixels on the screen without sacrificing security. Furthermore, we show that with the right type of cryptographic method, we can decrypt and display encrypted animated images at 30 FPS or higher for medium-to-small images and at around 30 FPS for large images. One notable cryptographic method we adapt for Rushmore is visual cryptography, and we demonstrate that it is a light-weight alternative to other cryptographic methods for certain use cases. Our evaluation shows that in addition to providing usable frame rates, Rushmore incurs less than 5% overhead to the applications running in the normal world.