MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Hello folks,

Suppose we're working in ${\mathbb T}^q$, the $q$-times Cartesian product of $[-\pi,\pi]$ with the endpoints identified. So, $ L^p := L^p({\mathbb T}^q) $.

The Young inequality may be written in the following form: For $p \in [1,\infty]$, $f \in L^p$, $g \in L^1$, we have $f * g \in L^p$ and

$\|f*g\|_p \le \|f\|_p \|g\|_1$.

Could we extend this to an arbitrary measure $\mu$ on ${\mathbb T}^q$ by:

Let $p \in [1,\infty]$, $f \in L^p$, $g \in L^1(\mu)$. Then $ f*_\mu g := \int_{{\mathbb T}^q} f(\circ-t)g(t)d\mu(t) \in L^p$ and $\|f*_\mu g\|_p \le \|f\|_p \|g\|_{\mu;1}$ ?

I'm working on a related problem in approximation theory with a barely-Rudin level understanding of measure theory, so I'm not at the level to prove it myself or find a counterexample, but the estimate (or one like it) is necessary to connect some dots in my head. Any inference as to why it should or shouldn't be true?

Sincere thanks for any help.

share|cite|improve this question
It's not true in general. One of the ways that Young's inequality is motivated (and essentially proved) is by using a scaling argument, which isn't possible for arbitrary measures. – Christopher A. Wong Mar 9 '12 at 16:45

The most general version of Young's convolution inequality that I know of is as follows: Let $(X,\mu)$ be a $\sigma$-finite measure space. For $1 \leq p, r \leq \infty$,

$\left\|f \ast g\right\|_{L^{s}} \leq \left\|f\right\|_{L^{r}}\left\|g\right\|_{L^{p}}$

where $\frac{1}{p} + \frac{1}{r} = 1 + \frac{1}{s}$.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.