Euclidean Preference

Euclidean preference models represent individual preferences as distances in a multi-dimensional space, assuming individuals prefer options closer to their "ideal point." Current research investigates the limitations of this model, focusing on the dimensionality required for accurate representation of various preference profiles using both Euclidean and Manhattan distance metrics. These studies explore the conditions under which a Euclidean representation is possible and quantify the error introduced when using lower-dimensional approximations, impacting the interpretation of vector embeddings in applications like recommender systems and multi-agent systems. Understanding these limitations is crucial for developing more robust and accurate preference modeling techniques.