Identifying the smallest dataset

"We study the fundamental question of how informative a dataset is for solving a given decision-making task. [pdf]

"In our setting, the dataset provides partial information about unknown parameters that influence task outcomes. 

"Focusing on linear programs, we characterize when a dataset is sufficient to recover an optimal decision, given an uncertainty set on the cost vector. 

"Our main contribution is a sharp geometric characterization that identifies the directions of the cost vector that matter for optimality, relative to the task constraints and uncertainty set. 

"We further develop a practical algorithm that, for a given task, constructs a minimal or least-costly sufficient dataset. 

"Our results reveal that small, well-chosen datasets can often fully determine optimal decisions —offering a principled foundation for task-aware data selection."



Comments

Popular posts from this blog

Hamza Chaudhry

Swarm 🦹‍♂️

Digital ID tracking system