no code implementations • 2 Jul 2019 • Alaa Maalouf, Adiel Statman, Dan Feldman
With high probability, non-uniform sampling based on upper bounds on what is known as importance or sensitivity of each row in $A$ yields a coreset.
