More Like this

1. Automated Task-Time Interventions to Improve Teamwork using Imitation Learning

   Seo, Sangwon; Han, Bing; Unhelkar, Vaibhav. (January 2023, AAMAS '23: Proceedings of the 2023 International Conference on Autonomous Agents and Multiagent Systems)

   Effective human-human and human-autonomy teamwork is critical but often challenging to perfect. The challenge is particularly relevant in time-critical domains, such as healthcare and disaster response, where the time pressures can make coordination increasingly difficult to achieve and the consequences of imperfect coordination can be severe. To improve teamwork in these and other domains, we present TIC: an automated intervention approach for improving coordination between team members. Using BTIL, a multi-agent imitation learning algorithm, our approach first learns a generative model of team behavior from past task execution data. Next, it utilizes the learned generative model and team's task objective (shared reward) to algorithmically generate execution-time interventions. We evaluate our approach in synthetic multi-agent teaming scenarios, where team members make decentralized decisions without full observability of the environment. The experiments demonstrate that the automated interventions can successfully improve team performance and shed light on the design of autonomous agents for improving teamwork. 

   more » « less
2. REFORM: Fast and Adaptive Solution for Subteam Replacement

   https://doi.org/10.1109/BigData52589.2021.9671359  

   Li, Zhaoheng; Pi, Xinyu; Wu, Mingyuan; Tong, Hanghang (December 2021, REFORM: Fast and Adaptive Solution for Subteam Replacement)

   Subteam Replacement: given a team of people em- bedded in a social network to complete a certain task, and a subset of members (i.e., subteam) in this team which have become unavailable, find another set of people who can perform the subteam’s role in the larger team. We conjecture that a good candidate subteam should have high skill and structural similarity with the replaced subteam while sharing a similar connection with the larger team as a whole. Based on this conjecture, we propose a novel graph kernel which evaluates the goodness of candidate subteams in this holistic way freely adjustable to the need of the situation. To tackle the significant computational difficulties, we equip our kernel with a fast approximation algorithm which (a) employs effective pruning strategies, (b) exploits the similarity between candidate team structures to reduce kernel computations, and (c) features a solid theoretical bound on the quality of the obtained solution. We extensively test our solution on both synthetic and real datasets to demonstrate its effectiveness and efficiency. Our proposed graph kernel outputs more human-agreeable recommendations compared to metrics used in previous work, and our algorithm consistently outperforms alternative choices by finding near- optimal solutions while scaling linearly with the size of the replaced subteam. 

   more » « less
3. Enhancing Teamwork Skills Through an Engineering Service-learning Collaboration

   Pazos, P.; Cima, F.; Kidd, J.; Ringleb, S.; & Ayala, O.; Gutierrez, K.; Kaipa, K. (June 2020, 2020 ASEE Virtual Annual Conference Content Access, Virtual Online)

   The purpose of this research paper is to explore whether participation in an interdisciplinary collaboration program partnering Preservice Teachers (PST) and Undergraduate Engineering Students (UES) results in an increase in teamwork effectiveness. The interdisciplinary collaboration was designed as a service-learning project within existing undergraduate programs that included the development and delivery of engineering content to a K-12 audience. The collaborations were integrated into existing courses in two colleges, engineering and education. The Behaviorally Anchored Rating Scale (BARS) version of the Comprehensive Assessment of Team Member Effectiveness (CATME) was used midway and at the end of the project to evaluate teamwork effectiveness. Results of the analysis indicated that both PST and UES experienced a significant increase in team-member effectiveness over the course of the project in four of the five factors: interacting with team members, keeping the team on track, expecting quality, and having relevant knowledge, skills and abilities. A noticeable positive increase in student attitudes towards the task was also observed between the midway and the end of the project. Analysis also suggests that the gain in the teamwork effectiveness did not differ across majors, with both UES and PST showing similar gains. Findings from this study provide some preliminary evidence that an innovative interdisciplinary service learning experience partnering engineering and education students, had a positive impact on their teamwork skills. 

   more » « less
4. Approaches to Evidencing Intra-Team Equity in Student Collaborative Design Decision-Making Interactions

   Moffat, A. D.; Fowler, R. R.; Matz, R. L.; JaQuay, S. (June 2023, American Society for Engineering Education)

   Team- and project-based pedagogies are increasingly normative in engineering education and beyond. Student teamwork holds the promise of developing collaborative skills deemed essential for new engineers by professional accreditation bodies such as ABET. The emphasis on these models, furthermore, reflects developments in pedagogical theory, stressing the importance of experiential learning and the social construction of knowledge, repositioning the instructor as a facilitator and guide. Teamwork in an educational context differs from that in professional contexts in that learning outcomes for all team members – both in terms of technical knowledge and team-working skills – are a primary goal of the activity, even while more tangible task-related outcomes might be the main concern of the students themselves. However, team-based learning also holds the potential for team members to have negative experiences, of which instructors may have little or no awareness, especially in real-time. Teams may achieve team-level outcomes required for successful completion, in spite of uneven levels of participation and contribution. Reduced participation on the part of an individual team member may have many causes, pro-active or reactive: it may be a deliberate refusal to engage, a lack of self-confidence, or a response to hostility from other members, among other possibilities. Inequitable team interactions will lead to uneven uptake of desired learning outcomes. Fostering equity in interactions and identifying inequitable practices among team members is therefore an important part of implementing team-based pedagogies, and an essential first step in identifying and challenging systematic patterns of inequity with regard to members of historically marginalized groups. This paper will therefore explore ways in which equity in group decision-making may be conceptualized and observed, laying the foundations for identifying and addressing inequities in the student experience. It will begin by considering different potential manifestations of interactional equity, surveying notions derived from prior education research in the fields of health, mathematics, engineering, and the natural sciences. These notions include: equity of participation on the basis of quantified vocal contributions (in terms of words, utterances, or clausal units); distribution and evolution of interactional roles; equity of idea endorsement and uptake; distribution of inchargeness and influence; equity of access to positional identities and discourse practices; and team member citizenship. In the paper’s empirical component, we trial measures of equity taken or developed from this literature on a small dataset of transcripts showing verbal interactions between undergraduate student team members in a first-year engineering design course. Some measures will be qualitative and others quantitative, depending on the particular form and manifestation of equity they are designed to examine. Measures include manual coding of speech acts and interactional ‘bids’, statistical measures of utterance frequency and length, and computational approaches to modeling interactional features such as social impact and receptivity. Results are compared with the students’ own reflections on the interactions, taken immediately afterward. Recommendations are made for the application of the measures, both from research and practice perspectives. Keywords: Teamwork, Equity, Interaction, Design 

   more » « less
5. Analyzing the Fluency of Human-Robot Interactions

   Weiss, Emily; Jacotin, Zeneve; Jackson, Ryan Blake; Yuan, Amy; Boerkoel, James (January 2023, AAAI 2023 Spring Symposium on HRI in Academia and Industry: Bridging the Gap)

   Fluency---described as the ``coordinated meshing of joint activities between members of a well-synchronized team''---is essential to human-robot team success. Human teams achieve fluency through rich, often mostly implicit, communication. A key challenge in bridging the gap between industry and academia is understanding what influences human perception of a fluent team experience to better optimize human-robot fluency in industrial environments. This paper addresses this challenge by developing an online experiment featuring videos that vary the timing of human and robot actions to influence perceived team fluency. Our results support three broad conclusions. First, we did not see differences across most subjective fluency measures. Second, people report interactions as more fluent as teammates stay more active. Third, reducing delays when humans' tasks depend on robots increases perceived team fluency. 

   more » « less