An official website of the United States government
Multiagent teams have been shown to be effective in many domains that require coordination among team members. However, finding valuable joint-actions becomes increasingly difficult in tightly-coupled domains where each agent’s performance depends on the actions of many other agents. Reward shaping partially addresses this challenge by deriving more “tuned" rewards to provide agents with additional feedback, but this approach still relies on agents ran- domly discovering suitable joint-actions. In this work, we introduce Counterfactual Agent Suggestions (CAS) as a method for injecting knowledge into an agent’s learning process within the confines of existing reward structures. We show that CAS enables agent teams to converge towards desired behaviors more reliably. We also show that improvement in team performance in the presence of suggestions extends to large teams and tightly-coupled domains.
more »
« less
Multidisciplinary learning through collective performance favors decentralization
Many models of learning in teams assume that team members can share solutions or learn concurrently. However, these assumptions break down in multidisciplinary teams where team members often complete distinct, interrelated pieces of larger tasks. Such contexts make it difficult for individuals to separate the performance effects of their own actions from the actions of interacting neighbors. In this work, we show that individuals can overcome this challenge by learning from network neighbors through mediating artifacts (like collective performance assessments). When neighbors’ actions influence collective outcomes, teams with different networks perform relatively similarly to one another. However, varying a team’s network can affect performance on tasks that weight individuals’ contributions by network properties. Consequently, when individuals innovate (through “exploring” searches), dense networks hurt performance slightly by increasing uncertainty. In contrast, dense networks moderately help performance when individuals refine their work (through “exploiting” searches) by efficiently finding local optima. We also find that decentralization improves team performance across a battery of 34 tasks. Our results offer design principles for multidisciplinary teams within which other forms of learning prove more difficult.
more »
« less
- Award ID(s):
- 2019470
- PAR ID:
- 10516507
- Publisher / Repository:
- Proceedings of the National Academy of Sciences
- Date Published:
- Journal Name:
- Proceedings of the National Academy of Sciences of the United States of America
- ISSN:
- 1091-6490
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
As online social networks (OSNs) become more prevalent, a new paradigm for problem-solving through crowd-sourcing has emerged. By leveraging the OSN platforms, users can post a problem to be solved and then form a team to collaborate and solve the problem. A common concern in OSNs is how to form effective collaborative teams, as various tasks are completed through online collaborative networks. A team’s diversity in expertise has received high attention to producing high team performance in developing team formation (TF) algorithms. However, the effect of team diversity on performance under different types of tasks has not been extensively studied. Another important issue is how to balance the need to preserve individuals’ privacy with the need to maximize performance through active collaboration, as these two goals may conflict with each other. This research has not been actively studied in the literature. In this work, we develop a team formation (TF) algorithm in the context of OSNs that can maximize team performance and preserve team members’ privacy under different types of tasks. Our proposedPRivAcy-Diversity-AwareTeamFormation framework, calledPRADA-TF, is based on trust relationships between users in OSNs where trust is measured based on a user’s expertise and privacy preference levels. The PRADA-TF algorithm considers the team members’ domain expertise, privacy preferences, and the team’s expertise diversity in the process of team formation. Our approach employs game-theoretic principlesMechanism Designto motivate self-interested individuals within a team formation context, positioning the mechanism designer as the pivotal team leader responsible for assembling the team. We use two real-world datasets (i.e., Netscience and IMDb) to generate different semi-synthetic datasets for constructing trust networks using a belief model (i.e., Subjective Logic) and identifying trustworthy users as candidate team members. We evaluate the effectiveness of our proposedPRADA-TFscheme in four variants against three baseline methods in the literature. Our analysis focuses on three performance metrics for studying OSNs: social welfare, privacy loss, and team diversity.more » « less
-
Ruis, Andrew R.; Lee, Seung B. (Ed.)Quantitative ethnographers across a range of domains study complex collaborative thinking (CCT): the processes by which members of a group or team develop shared understanding by making cognitive connections from the statements and actions of the group. CCT is difficult to model because the actions of group members are interdependent—the activity of any individual is influenced by the actions of other members of the group. Moreover, the actions of group members engaged in some collaborative tasks may need to follow a particular order. However, current techniques can account for either interdependence or order, but not both. In this paper, we present directed epistemic network analysis (dENA), an extension of epistemic network analysis (ENA), as a method that can simultaneously account for the interdependent and ordered aspects of CCT. To illustrate the method, we compare a qualitative analysis of two U.S. Navy commanders working in a simulation to ENA and dENA analyses of their performance. We find that by accounting for interdependence but not order, ENA was not able to model differences between the commanders seen in the qualitative analysis, but by accounting for both interdependence and order, dENA was able to do so.more » « less
-
Silva, S; Paquete, L (Ed.)Coevolving teams of agents promises effective solutions for many coordination tasks such as search and rescue missions or deep ocean exploration. Good team performance in such domains generally relies on agents discovering complex joint policies, which is particularly difficult when the fitness functions are sparse (where many joint policies return the same or even zero fitness values). In this paper, we introduce Novelty Seeking Multiagent Evolutionary Reinforcement Learning (NS-MERL), which enables agents to more efficiently explore their joint strategy space. The key insight of NS-MERL is to promote good exploratory behaviors for individual agents using a dense, novelty-based fitness function. Though the overall team-level performance is still evaluated via a sparse fitness function, agents using NS-MERL more efficiently explore their joint action space and more readily discover good joint policies. Our results in complex coordination tasks show that teams of agents trained with NS-MERL perform significantly better than agents trained solely with task-specific fitnesses.more » « less
-
This research paper investigates how individual change agents come together to form effective teams. Improving equity within academic engineering requires changes that are often too complex and too high-risk for a faculty member to pursue on their own. Teams offer the advantage of combining a diverse skill set of many individuals, as well as bringing together insider knowledge and external specialist expertise. However, in order for teams of academic change agents to function effectively, they must overcome the challenges of internal politics, power differentials, and group conflict. This analysis of team formation emerges from our participatory action research with recipients of the NSF Revolutionizing Engineering Departments (RED) grants. Through an NSF-funded collaboration between the University of Washington and Rose-Hulman Institute of Technoliogy, we work with the RED teams to research the process of change as they work to improve equity and inclusion within their institutions. Utilizing longitudinal qualitative data from focus group discussions with 16 teams at the beginning and midpoints of their projects, we examine the development of teams to transform engineering education. Drawing on theoretical frameworks from social movement theory, we highlight the importance of creating a unified team voice and developing a sense of group agency. Teams have a better chance of achieving their goals if members are able to create a unified voice—that is, a shared sense of purpose and vision for their team. We find that the development of a team’s unified voice begins with proposal writing. When members of RED teams did not collaboratively write the grant proposal, they found it necessary to devote more time to develop a sense of shared vision for their project. For many RED teams, the development of a unified voice was further strengthened through external messaging, as they articulated a “we” in opposition to a “they” who have different values or interests. Group agency develops as a result of team members perceiving their goals as attainable and their efforts, as both individuals and a group, as worthwhile. That is, group agency is dependent on both the credibility of the team as well as trust among team members. For some of the RED teams, the NSF requirement to include social scientists and education researchers on their teams gave the engineering team members new, increased exposure to these fields. RED teams found that creating mutual respect was foundational for working across disciplinary differences and developing group agency.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1073/pnas.2303568120
