Dyadic Regression Model

Dyadic regression models analyze relationships between pairs of entities, predicting outcomes based on their interaction. Current research focuses on improving model accuracy and fairness, addressing biases revealed by novel evaluation metrics like Eccentricity-Area Under the Curve (EAUC), and exploring diverse applications such as human-robot interaction and personalized medicine. These models are increasingly important for understanding complex interactions in various fields, with ongoing work developing algorithms like dyadic reinforcement learning and leveraging techniques such as deep learning and natural language processing for improved prediction and explainability.