Team Synchronization and Individual Contributions in Coop-Space Fortress

Li, Huao; Ni, Tianwei; Agrawal, Siddharth; Hughes, Dana; Sycara, Katia

doi:10.1177/1071181320641023

Cited by 5 publications

(8 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, we proposed a novel adaptive agent framework in human agent teaming (HAT) based on the cross-entropy similarity measure and a pre-trained static policy library. The framework was inspired by human teamwork research, which illustrates important characteristics of teamwork such as the existence of complementary policies, influence of adaptive actions on team performance, and the dynamic human policies in cooperation [23], [24]. The CEM measure making it possible to directly measure and compare behavior among humans and agents provides a tool of broad applicability.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Individualized Mutual Adaptation in Human-Agent Teams

Agrawal

et al. 2021

IEEE Trans. Human-Mach. Syst.

Self Cite

View full text Add to dashboard Cite

The ability to collaborate with previously unseen human teammates is crucial for artificial agents to be effective in human-agent teams (HATs). Due to individual differences and complex team dynamics, it is hard to develop a single agent policy to match all potential teammates. In this paper, we study both human-human and humanagent teams in a dyadic cooperative task, Team Space Fortress (TSF). Results show that the team performance is influenced by both players' individual skill level and their ability to collaborate with different teammates by adopting complementary policies. Based on human-human team results, we propose an adaptive agent that identifies different human policies and assigns a complementary partner policy to optimize team performance. The adaptation method relies on a novel similarity metric to infer human policy and then selects the most complementary policy from a pre-trained library of exemplar policies. We conducted human-agent experiments to evaluate the adaptive agent and examine mutual adaptation in humanagent teams. Results show that both human adaptation and agent adaptation contribute to team performance.

show abstract

Section: Discussionmentioning

confidence: 99%

“…Analyzing human teamwork also gives us a baseline when designing agents in terms of reward function and adaptive behaviors. The results of this human teamwork experiment are reported in more detail in [23], [24]. Here we briefly discuss the insights it brings to our work in human agent teaming.…”

Section: Human-human Teamsmentioning

confidence: 99%

Individualized Mutual Adaptation in Human-Agent Teams

Agrawal

et al. 2021

IEEE Trans. Human-Mach. Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

“…In recent work, Li et al (2020) examined teammate collaboration and action adaptation during the mission in tightly coupled, time stressed teamwork. In collaborative tasks, the final outcome is determined by individual level (e.g., skill level, motivation, and personality) and team level (e.g., communication, team coherence, and complementary strategy) criterion of interest.…”

Section: Human Teammate Substitutementioning

confidence: 99%

“…Identifying teammate contributions is complicated by differences in play due to adaptation to other teammates leading to equilibria in which each player's actions and contributions may be quite distinct from team to team. In this work (Li et al, 2020), the researchers identified the components of team adaptation in more detail, with respect to not only the final outcome (performance measure) but also the adaptive behaviors that occur as the mission evolves, a quantification that is novel in the human factors and multi-agent literature. Space Fortress (Agarwal et al, 2019), a game that has been used extensively for psychological research, was adapted for investigate teams.…”

Section: Human Teammate Substitutementioning

confidence: 99%

Human–Autonomy Teaming: Definitions, Debates, and Directions

et al. 2021

Self Cite

View full text Add to dashboard Cite

Researchers are beginning to transition from studying human–automation interaction to human–autonomy teaming. This distinction has been highlighted in recent literature, and theoretical reasons why the psychological experience of humans interacting with autonomy may vary and affect subsequent collaboration outcomes are beginning to emerge (de Visser et al., 2018; Wynne and Lyons, 2018). In this review, we do a deep dive into human–autonomy teams (HATs) by explaining the differences between automation and autonomy and by reviewing the domain of human–human teaming to make inferences for HATs. We examine the domain of human–human teaming to extrapolate a few core factors that could have relevance for HATs. Notably, these factors involve critical social elements within teams that are central (as argued in this review) for HATs. We conclude by highlighting some research gaps that researchers should strive toward answering, which will ultimately facilitate a more nuanced and complete understanding of HATs in a variety of real-world contexts.

show abstract

“…Before constructing exemplar policies, we first evaluate the nature of this testbed through previous research in human-human teams. The results (Li et al 2020b) show that different human-human pairs demonstrate significantly diverse performance and the team performance was affected by both individual level factors such as skill levels and team level factors such as team synchronization and adaptation.…”

Section: Introductionmentioning

confidence: 98%

Adaptive Agent Architecture for Real-time Human-Agent Teaming

Ni,

Li,

Agrawal

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Teamwork is a set of interrelated reasoning, actions and behaviors of team members that facilitate common objectives. Teamwork theory and experiments have resulted in a set of states and processes for team effectiveness in both humanhuman and agent-agent teams. However, human-agent teaming is less well studied because it is so new and involves asymmetry in policy and intent not present in human teams. To optimize team performance in human-agent teaming, it is critical that agents infer human intent and adapt their polices for smooth coordination. Most literature in human-agent teaming builds agents referencing a learned human model. Though these agents are guaranteed to perform well with the learned model, they lay heavy assumptions on human policy such as optimality and consistency, which is unlikely in many real-world scenarios. In this paper, we propose a novel adaptive agent architecture in human-model-free setting on a twoplayer cooperative game, namely Team Space Fortress (TSF). Previous human-human team research have shown complementary policies in TSF game and diversity in human players' skill, which encourages us to relax the assumptions on human policy. Therefore, we discard learning human models from human data, and instead use an adaptation strategy on a pre-trained library of exemplar policies composed of RL algorithms or rule-based methods with minimal assumptions of human behavior. The adaptation strategy relies on a novel similarity metric to infer human policy and then selects the most complementary policy in our library to maximize the team performance. The adaptive agent architecture can be deployed in real-time and generalize to any off-the-shelf static agents. We conducted human-agent experiments to evaluate the proposed adaptive agent framework, and demonstrated the suboptimality, diversity, and adaptability of human policies in human-agent teams.* indicates equal contribution. In AAAI 2021 Workshop on Plan, Activity, and Intent Recognition.

show abstract

Team Synchronization and Individual Contributions in Coop-Space Fortress

Cited by 5 publications

References 14 publications

Individualized Mutual Adaptation in Human-Agent Teams

Individualized Mutual Adaptation in Human-Agent Teams

Human–Autonomy Teaming: Definitions, Debates, and Directions

Adaptive Agent Architecture for Real-time Human-Agent Teaming

Contact Info

Product

Resources

About