Partner Capability Estimation for Task-Agnostic Adaptation in Ad-Hoc Teamwork
2026-07-29 • Artificial Intelligence
Artificial IntelligenceHuman-Computer InteractionMultiagent Systems
AI summaryⓘ
The authors focus on improving teamwork between autonomous agents and new, unknown partners, especially when partners' abilities are not clear and tasks can be varied. They create a method called CE-CM that estimates partner capabilities without needing lots of prior training and updates its understanding quickly by observing a few tasks. They also introduce CE-CM-Div, which handles the unpredictability of human behavior by considering multiple possible actions rather than just the best one. Their experiments show these methods help agents better understand and adapt to partners over time, which is important for effective human-AI collaboration.
ad-hoc teamworkautonomous agentscapability estimationBayesian inferencemulti-task learningMarkov Decision Processesjoint planninghuman-AI teamingsimulation-based samplingbehavioral diversity
Authors
Peter Tisnikar, Maja Swieczkowska, Benteng Ma, Gerard Canal, Matteo Leonetti
Abstract
Effective collaboration with novel and diverse partners is a crucial skill for autonomous agents. Most current ad-hoc teamwork (AHT) approaches assume that agents will collaborate on a single, fixed task and that the partner's capabilities, their ability to successfully execute the desired action, are already known. In reality, a partner's true capabilities are often hidden, and human collaborators may act sub-optimally on tasks with multiple valid strategies. To address these limitations, we extend ad-hoc teamwork into a multi-task setting by re-framing it as a problem of joint planning with decentralised execution under hidden partner capabilities. We introduce CE-CM (Capability Estimation via Contextual Models), an approximate Bayesian method that infers task-invariant capability vectors. By using simulation-based sampling, the agent estimates capabilities and induces a contextual Multi-agent Markov Decision Processes for planning. This approach requires no population pre-training and refines its beliefs online from just a few tasks. To account for human unpredictability, we propose CE-CM-Div, an extension that evaluates capability hypotheses against diverse planner rollouts rather than a single optimal trajectory. Simulated experiments demonstrate that CE-CM rapidly recovers hidden capabilities, reduces infeasible action assignments, and adapts to changes over time. Furthermore, in an offline human study of 225 trajectories from 15 participants, CE-CM-Div substantially improved capability estimates over the baseline CE-CM method. Our results suggest capability-based modelling is a promising interpretable, task-agnostic representation in the studied settings, demonstrating that accounting for behavioural diversity is essential for robust human-AI teaming.