The Kato problem, or Kato square root problem, asked whether the square root of a uniformly elliptic divergence-form operator
on has the same domain
as the first-order gradient. The space
is the L2-space
of equivalence classes of square
integrable functions
with respect to Lebesgue measure, meaning
. When
, the notation is
. Here
is a bounded matrix-valued measurable
function satisfying uniform ellipticity.
The answer is affirmative: the domain of is the Sobolev space
, and
for . The problem originated
in Kato's (1961) question whether a regularly accretive operator
and its adjoint have square
roots with the same domain. Auscher et al. (2002)
proved the divergence-form result above in all dimensions.