More Like this

1. Co-designing for Education

   Bettencourt, Benjamin; Pera, Maria_Soledad; Kennington, Casey; Landau_Wright, Katherine; Fails, Jerry_Alan (June 2024, IDC 2024 Technology-Enhanced STEM Learning in Childhood Workshop)

   Designing for child-learning involves a number of stakeholders including, but not limited to: teachers, children, administrators, and families. A common approach used to design technologies is co-design. Yet, co-design frequently means different things for different stakeholders. Within the realm of education co-design can be used generally for any interaction with a stakeholder that can be used to guide or inform the design of the desired outcome (product or curriculum) -- often with different stakeholders separately and/or in very small groups (e.g. a group of teachers or 2-3 children, or a classroom if ``testing''). Within the field of child-computer interaction, designing technologies with and for children can involve children and other stakeholders in varying levels of involvement, although within the IDC community it is often a more substantial contribution. We posit that giving child stakeholders an authentic voice in the design of technologies is crucial to fully addressing stakeholder's needs. 

   more » « less
2. Strategy for the Co-Assembly of Co-Axial Nanotube–Polymer Hybrids

   https://doi.org/10.1021/acs.langmuir.7b02245  

   Ji, Mingyang; Dawadi, Mahesh B.; LaSalla, Alexandria R.; Sun, Yuan; Modarelli, David A.; Parquette, Jon R. (August 2017, Langmuir)
3. Co-Designing Situated Displays for Family Co-Regulation with ADHD Children

   https://doi.org/10.1145/3613904.3642745  

   Silva, Lucas M; Cibrian, Franceli L; Bonang, Clarisse; Bhattacharya, Arpita; Min, Aehong; Monteiro, Elissa M; Beltran, Jesus Armando; Schuck, Sabrina; Lakes, Kimberley D; Hayes, Gillian R; et al (May 2024, ACM)

   Family informatics often uses shared data dashboards to promote awareness of each other’s health-related behaviors. However, these interfaces often stop short of providing families with needed guidance around how to improve family functioning and health behaviors. We consider the needs of family co-regulation with ADHD children to understand how in-home displays can support family well-being. We conducted three co-design sessions with each of eight families with ADHD children who had used a smartwatch for self-tracking. Results indicate that situated displays could nudge families to jointly use their data for learning and skill-building. Accommodating individual needs and preferences when family members are alone is also important, particularly to support parents exploring their co-regulation role, and assisting children with data interpretation and guidance on self and co-regulation. We discuss opportunities for displays to nurture multiple intents of use, such as joint or independent use, while potentially connecting with external expertise. 

   more » « less
4. Co-training for Policy Learning

   song, J (July 2019, Proceedings of The 35th Uncertainty in Artificial Intelligence Conference)

   null (Ed.)

   We study the problem of learning sequential decision-making policies in settings with multiple state-action representations. Such settings naturally arise in many domains, such as planning (e.g., multiple integer programming formulations) and various combinatorial optimization problems (e.g., those with both integer programming and graph-based formulations). Inspired by the classical co-training framework for classification, we study the problem of co-training for policy learning. We present sufficient conditions under which learning from two views can improve upon learning from a single view alone. Motivated by these theoretical insights, we present a meta-algorithm for co-training for sequential decision making. Our framework is compatible with both reinforcement learning and imitation learning. We validate the effectiveness of our approach across a wide range of tasks, including discrete/continuous control and combinatorial optimization. 

   more » « less
5. Co-training for Policy Learning

   Song, JL; Lanka, RA; Yue, Y; Ono, MSRO (January 2020, 35TH UNCERTAINTY IN ARTIFICIAL INTELLIGENCE CONFERENCE (UAI 2019))

   Adams, RP; Gogate V (Ed.)

   We study the problem of learning sequential decision-making policies in settings with multiple state-action representations. Such settings naturally arise in many domains, such as planning (e.g., multiple integer programming formulations) and various combinatorial optimization problems (e.g., those with both integer programming and graph-based formulations). Inspired by the classical co-training framework for classification, we study the problem of co-training for policy learning. We present sufficient conditions under which learning from two views can improve upon learning from a single view alone. Motivated by these theoretical insights, we present a meta-algorithm for co-training for sequential decision making. Our framework is compatible with both reinforcement learning and imitation learning. We validate the effectiveness of our approach across a wide range of tasks, including discrete/continuous control and combinatorial optimization. 

   more » « less