1: \begin{abstract}

2: By a result of Schur [\textit{J.~reine angew.~Math.}~1911], the entrywise

3: product $M \circ N$ of two positive semidefinite matrices $M,N$ is again

4: positive.

5: Vyb\'iral [\textit{Adv.\ Math.}~2020] improved on this by showing the

6: uniform lower bound $M \circ \overline{M} \geq E_n / n$ for all $n \times

7: n$ real or complex correlation matrices $M$, where $E_n$ is the all-ones

8: matrix.

9: This was applied to settle a conjecture of Novak [\textit{J.~Complexity}

10: 1999] and to positive definite functions on groups.

11: Vyb\'iral (in his original preprint) asked if one can obtain similar

12: uniform lower bounds for higher entrywise powers of $M$, or for $M \circ

13: N$ when $N \neq M, \overline{M}$. A natural third question is to ask for

14: a tighter lower bound that does not vanish as $n \to \infty$, i.e.~over

15: infinite-dimensional Hilbert spaces.

16: 

17: In this note, we affirmatively answer all three questions by extending

18: and refining Vyb\'iral's result to lower-bound $M \circ N$, for arbitrary

19: complex positive semidefinite matrices $M,N$.

20: Specifically: we provide tight lower bounds, improving on Vyb\'iral's

21: bounds. Second, our proof is `conceptual' (and self-contained), providing

22: a natural interpretation of these improved bounds via tracial

23: Cauchy--Schwarz inequalities. Third, we extend our tight lower bounds to

24: Hilbert--Schmidt operators.

25: As an application, we settle Open Problem~1 of

26: Hinrichs--Krieg--Novak--Vyb\'iral [\textit{J.\ Complexity}, in press],

27: which yields improvements in the error bounds in certain tensor product

28: (integration) problems.

29: \end{abstract}

30: