Gennie Mansi

Gennie Mansi, Ashley Boone, Sue Reon Kim et al.

Art education plays a significant role in K-2 learners' physical and cognitive development. However, teachers struggle to translate in-person activities to remote settings and to give necessary feedback to help learners develop fine motor skills. Previous research shows the benefits of tangible technology and real-time system feedback for supporting teachers and students in digital environments, but little research explores their affordances for remote art education. We developed Chameleon Clippers: interactive scissors that give real-time feedback to learners as they cut along a line. In preliminary tests, learners felt engaged and responded to feedback, enjoying their experience. Our low-cost design augments existing classroom artifacts and practices, supporting classroom integration. Testing also revealed directions for future study, including the frequency of feedback and assimilation into a broader, art education platform. Through our study, we demonstrate the potential for tangible technology to create more interactive, engaging, and supportive remote K-2 learning experiences.

Amal Alabdulkarim, Madhuri Singh, Gennie Mansi et al. · gatech

Reinforcement learning (RL) systems can be complex and non-interpretable, making it challenging for non-AI experts to understand or intervene in their decisions. This is due in part to the sequential nature of RL in which actions are chosen because of their likelihood of obtaining future rewards. However, RL agents discard the qualitative features of their training, making it difficult to recover user-understandable information for "why" an action is chosen. We propose a technique Experiential Explanations to generate counterfactual explanations by training influence predictors along with the RL policy. Influence predictors are models that learn how different sources of reward affect the agent in different states, thus restoring information about how the policy reflects the environment. Two human evaluation studies revealed that participants presented with Experiential Explanations were better able to correctly guess what an agent would do than those presented with other standard types of explanation. Participants also found that Experiential Explanations are more understandable, satisfying, complete, useful, and accurate. Qualitative analysis provides information on the factors of Experiential Explanations that are most useful and the desired characteristics that participants seek from the explanations.

Madhuri Singh, Gennie Mansi, Mark Owen Riedl

K-12 teachers employ Engineering Design Challenges to help students learn about the Engineering Design Process hands-on. They use techniques like hard scaffolding questions to guide the students as they think through the different stages of the engineering design process. While useful, the creation of these questions adds to the teacher's preparation time for their classes. Concept Catalyst uses Large Language Models to assist teachers with the rapid creation of scaffold questions for engineering design challenges. Unlike open-ended chat, Concept Catalyst uses LLMs to summarize and decompose an engineering design challenge into the concepts that students will engage with, allow the teacher to visually manipulate and link related concepts, and to propose scaffolding questions for the teacher to modify or accept.

Madhuri Singh, Amal Alabdulkarim, Gennie Mansi et al. · gatech

Explainable AI (XAI) systems have been proposed to help people understand how AI systems produce outputs and behaviors. Explainable Reinforcement Learning (XRL) has an added complexity due to the temporal nature of sequential decision-making. Further, non-AI experts do not necessarily have the ability to alter an agent or its policy. We introduce a technique for using World Models to generate explanations for Model-Based Deep RL agents. World Models predict how the world will change when actions are performed, allowing for the generation of counterfactual trajectories. However, identifying what a user wanted the agent to do is not enough to understand why the agent did something else. We augment Model-Based RL agents with a Reverse World Model, which predicts what the state of the world should have been for the agent to prefer a given counterfactual action. We show that explanations that show users what the world should have been like significantly increase their understanding of the agent policy. We hypothesize that our explanations can help users learn how to control the agents execution through by manipulating the environment.

Gennie Mansi, Mark Riedl

A core assumption of explainable AI systems is that explanations change what users know, thereby enabling them to act within their complex socio-technical environments. Despite the centrality of action, explanations are often organized and evaluated based on technical aspects. Prior work varies widely in the connections it traces between information provided in explanations and resulting user actions. An important first step in centering action in evaluations is understanding what the XAI community collectively recognizes as the range of information that explanations can present and what actions are associated with them. In this paper, we present our framework, which maps prior work on information presented in explanations and user action, and we discuss the gaps we uncovered about the information presented to users.