Intelligent behaviour in the physical world exhibits structure at multiple spatial and temporal scales. Although movements are ultimately executed at the level of instantaneous muscle tensions or joint torques, they must be selected to serve goals defined on much longer timescales, and in terms of relations that extend far beyond the body itself, ultimately involving coordination with other agents. Recent research in artificial intelligence has shown the promise of learning-based approaches to the respective problems of complex movement, longer-term planning and multi-agent coordination. However, there is limited research aimed at their integration. We study this problem by training teams of physically simulated humanoid avatars to play football in a realistic virtual environment. We develop a method that combines imitation learning, single- and multi-agent reinforcement learning and population-based training, and makes use of transferable representations of behaviour for decision making at different levels of abstraction. In a sequence of stages, players first learn to control a fully articulated body to perform realistic, human-like movements such as running and turning; they then acquire mid-level football skills such as dribbling and shooting; finally, they develop awareness of others and play as a team, bridging the gap between low-level motor control at a timescale of milliseconds, and coordinated goal-directed behaviour as a team at the timescale of tens of seconds. We investigate the emergence of behaviours at different levels of abstraction, as well as the representations that underlie these behaviours using several analysis techniques, including statistics from real-world sports analytics. Our work constitutes a complete demonstration of integrated decision-making at multiple scales in a physically embodied multi-agent setting. See project video at https://youtu.be/KHMwq9pv7mg.

This paper proposes a novel integrated dynamic method based on Behavior Trees for planning and allocating tasks in mixed human robot teams, suitable for manufacturing environments. The Behavior Tree formulation allows encoding a single job as a compound of different tasks with temporal and logic constraints. In this way, instead of the well-studied offline centralized optimization problem, the role allocation problem is solved with multiple simplified online optimization sub-problem, without complex and cross-schedule task dependencies. These sub-problems are defined as Mixed-Integer Linear Programs, that, according to the worker-actions related costs and the workers' availability, allocate the yet-to-execute tasks among the available workers. To characterize the behavior of the developed method, we opted to perform different simulation experiments in which the results of the action-worker allocation and computational complexity are evaluated. The obtained results, due to the nature of the algorithm and to the possibility of simulating the agents' behavior, should describe well also how the algorithm performs in real experiments.

Autonomous discovery and direct instruction are two extreme sources of learning in children, but educational sciences have shown that intermediate approaches such as assisted discovery or guided play resulted in better acquisition of skills. When turning to Artificial Intelligence, the above dichotomy is translated into the distinction between autonomous agents which learn in isolation and interactive learning agents which can be taught by social partners but generally lack autonomy. In between should stand teachable autonomous agents: agents learning from both internal and teaching signals to benefit from the higher efficiency of assisted discovery. Such agents could learn on their own in the real world, but non-expert users could drive their learning behavior towards their expectations. More fundamentally, combining both capabilities might also be a key step towards general intelligence. In this paper we elucidate obstacles along this research line. First, we build on a seminal work of Bruner to extract relevant features of the assisted discovery processes. Second, we describe current research on autotelic agents, i.e. agents equipped with forms of intrinsic motivations that enable them to represent, self-generate and pursue their own goals. We argue that autotelic capabilities are paving the way towards teachable and autonomous agents. Finally, we adopt a social learning perspective on tutoring interactions and we highlight some components that are currently missing to autotelic agents before they can be taught by ordinary people using natural pedagogy, and we provide a list of specific research questions that emerge from this perspective.

Recently, deep reinforcement learning (RL) algorithms have made great progress in multi-agent domain. However, due to characteristics of RL, training for complex tasks would be resource-intensive and time-consuming. To meet this challenge, mutual learning strategy between homogeneous agents is essential, which is under-explored in previous studies, because most existing methods do not consider to use the knowledge of agent models. In this paper, we present an adaptation method of the majority of multi-agent reinforcement learning (MARL) algorithms called "KnowSR" which takes advantage of the differences in learning between agents. We employ the idea of knowledge distillation (KD) to share knowledge among agents to shorten the training phase. To empirically demonstrate the robustness and effectiveness of KnowSR, we performed extensive experiments on state-of-the-art MARL algorithms in collaborative and competitive scenarios. The results demonstrate that KnowSR outperforms recently reported methodologies, emphasizing the importance of the proposed knowledge sharing for MARL.

We study the classic problem of fairly allocating a set of indivisible goods among a group of agents, and focus on the notion of approximate proportionality known as PROPm. Prior work showed that there exists an allocation that satisfies this notion of fairness for instances involving up to five agents, but fell short of proving that this is true in general. We extend this result to show that a PROPm allocation is guaranteed to exist for all instances, independent of the number of agents or goods. Our proof is constructive, providing an algorithm that computes such an allocation and, unlike prior work, the running time of this algorithm is polynomial in both the number of agents and the number of goods.

Reinforcement learning (RL) is a general framework that allows systems to learn autonomously through trial-and-error interaction with their environment. In recent years combining RL with expressive, high-capacity neural network models has led to impressive performance in a diverse range of domains. However, dealing with the large state and action spaces often required for problems in the real world still remains a significant challenge. In this paper we introduce a new simulation environment, "Gambit", designed as a tool to build scenarios that can drive RL research in a direction useful for military analysis. Using this environment we focus on an abstracted and simplified room clearance scenario, where a team of blue agents have to make their way through a building and ensure that all rooms are cleared of (and remain clear) of enemy red agents. We implement a multi-agent version of feudal hierarchical RL that introduces a command hierarchy where a commander at the higher level sends orders to multiple agents at the lower level who simply have to learn to follow these orders. We find that breaking the task down in this way allows us to solve a number of non-trivial floorplans that require the coordination of multiple agents much more efficiently than the standard baseline RL algorithms we compare with. We then go on to explore how qualitatively different behaviour can emerge depending on what we prioritise in the agent's reward function (e.g. clearing the building quickly vs. prioritising rescuing civilians).

We study the fundamental problem of allocating indivisible goods to agents with additive preferences. We consider eliciting from each agent only a ranking of her $k$ most preferred goods instead of her full cardinal valuations. We characterize the value of $k$ needed to achieve envy-freeness up to one good and approximate maximin share guarantee, two widely studied fairness notions. We also analyze the multiplicative loss in social welfare incurred due to the lack of full information with and without the fairness requirements.

We introduce the Cooperative Multi-Agent Path Finding (Co-MAPF) problem, an extension to the classical MAPF problem, where cooperative behavior is incorporated. In this setting, a group of autonomous agents operate in a shared environment and have to complete cooperative tasks while avoiding collisions with the other agents in the group. This extension naturally models many real-world applications, where groups of agents are required to collaborate in order to complete a given task. To this end, we formalize the Co-MAPF problem and introduce Cooperative Conflict-Based Search (Co-CBS), a CBS-based algorithm for solving the problem optimally for a wide set of Co-MAPF problems. Co-CBS uses a cooperation-planning module integrated into CBS such that cooperation planning is decoupled from path planning. Finally, we present empirical results on several MAPF benchmarks demonstrating our algorithm's properties.

The Nearly Autonomous Management and Control System (NAMAC) is a comprehensive control system that assists plant operations by furnishing control recommendations to operators in a broad class of situations. This study refines a NAMAC system for making reasonable recommendations during complex loss-of-flow scenarios with a validated Experimental Breeder Reactor II simulator, digital twins improved by machine-learning algorithms, a multi-attribute decision-making scheme, and a discrepancy checker for identifying unexpected recommendation effects. We assessed the performance of each NAMAC component, while we demonstrated and evaluated the capability of NAMAC in a class of loss-of-flow scenarios.

There are many unknowns regarding the characteristics and dynamics of human-AI teams, including a lack of understanding of how certain human-human teaming concepts may or may not apply to human-AI teams and how this composition affects team performance. This paper outlines an experimental research study that investigates essential aspects of human-AI teaming such as team performance, team situation awareness, and perceived team cognition in various mixed composition teams (human-only, human-human-AI, human-AI-AI, and AI-only) through a simulated emergency response management scenario. Results indicate dichotomous outcomes regarding perceived team cognition and performance metrics, as perceived team cognition was not predictive of performance. Performance metrics like team situational awareness and team score showed that teams composed of all human participants performed at a lower level than mixed human-AI teams, with the AI-only teams attaining the highest performance. Perceived team cognition was highest in human-only teams, with mixed composition teams reporting perceived team cognition 58% below the all-human teams. These results inform future mixed teams of the potential performance gains in utilizing mixed teams' over human-only teams in certain applications, while also highlighting mixed teams' adverse effects on perceived team cognition.