Loading...
SCI PhD Lunch Talk Series: Paras Sharma and Kuheli Sai
Private Location (sign in to display)
Details
Registration for this event is open to all graduate students at SCI. Join us to explore their research, share insights, and enjoy a complimentary lunch!
The following students shall be presenting. Details for their talks are as follows:
Paras Sharma
Title: Beyond Static Measures: Temporal Analysis of Lexical Alignment in Human-Human Learning With a Teachable RobotAbstract: Lexical alignment occurs when conversational partners converge on similar linguistic patterns. In collaborative learning settings, lexical alignment could indicate rapport, which can further predict learning and the collaborators' evolving shared understanding. Traditional approaches to alignment computation often focus more on the summary statistics computed at the end of the conversation, which usually do not capture the conversational dynamics efficiently. This work investigates how alignment evolves in a conversation by modeling lexical alignment trajectories between human dyads while interacting with a teachable robot. We find that along with the summary statistics, the alignment curve parameters and the time taken to reach key alignment moments significantly predict rapport. We further see significant relationships between the early turns in the conversation and the overall alignment trajectories, indicating the importance of modeling conversational dynamics to plan real-time interventions in the robot, altering the alignment trajectories, and, consequently learning outcomes.
Bio: Paras is a 3rd year Ph.D. student in the Computer Science department at the University of Pittsburgh working with Dr. Erin Walker in FACET Lab. His research interests lie at the intersection of Human-Computer Interaction, Natural Language Processing, and Multimodal Machine Learning particularly focusing on building educational technologies to help learners navigate through open-ended learning environments. He is interested in modeling learner behaviors during their multimodal interactions with educational systems and then utilizing these models to support varied learner dialogue interactions within the systems. Personal Website
Kuheli Sai
Title: Data-Driven and Sustainable Approach to Bridge Digital DivideAbstract: Next-Generation (NextG) wireless networks could widen the digital divide if not deployed equitably. Without targeted intervention, intentional planning, and provisioning of services from the network service providers and government, underserved and marginalized communities could be left behind, exacerbating pre-existing societal inequalities further in the digital domain. To address this societal problem, we propose a data-driven solution that prioritizes the installation of micro edge data centers, a key component of NextG networks, in socio-economically distressed neighborhoods within an urban area. In this talk, I will discuss and elaborate on how we can leverage concepts from data science, machine learning, and data from open data sources towards bridging NextG induced digital divide. Additionally, I will discuss how we can simultaneously achieve sustainability.
Bio: Kuheli Sai is a researcher in the domain of Data Science for Societal Good, and a Ph.D. Candidate in Information Science at the University of Pittsburgh. She is interested in tackling societal, environmental, public sector, and technological problems that help make progress towards achieving sustainable development goals. She uses concepts from data science and machine learning and primarily adopts empirical methods that specifically aim to solve social good problems. Her research simultaneously has implications for contributing towards policy development. Further details can be accessed from her website at: https://sites.google.com/view/kuhelisai
Food Provided