A continuous embedding is an injective continuous linear inclusion, so for a fixed constant. On a finite-measure domain, Holder inequality giveswith . For finite apply Holder inequality to with exponents and ; equality of exponents is immediate, and follows by integrating the supremum bound.
Weak convergence means for every bounded linear functional . The Riemann-Lebesgue lemma says that the Fourier transform of an integrable function tends to zero at high frequency. Since , it makes weakly null in . Its absolute value is at most one, so . For a test ,while by the Riemann-Lebesgue lemma applied to . Hence . Taking instead gives , so two weakly null factors need not have a weakly null product.
For and , Holder inequality givesThus in . In contrast, if had a weak limit , testing on every interval strictly away from zero would give integral zero for , hence almost everywhere by the permitted fact. Testing against the constant function one would give , a contradiction. Therefore has no weak limit.
In three dimensions let have compact support. Let solve and extend it by zero. There is a unique with for all . Indeed Holder inequality and the homogeneous Sobolev inequality make the right side a continuous linear functional in the gradient norm, and the Riesz representation theorem applies to this Hilbert space.
The weak formulation says that is the orthogonal projection of onto the closed subspace consisting of zero extensions of . These subspaces increase with and their union contains every test function, which is dense by the definition of the homogeneous Sobolev space. The best-approximation property of orthogonal projection therefore gives . The homogeneous Sobolev inequality also gives . This convergence does not generally extend to the inhomogeneous Sobolev space , as the spherical mean of a Poisson solution outside its source shows for a source of nonzero total mass.
Slow-rate prediction bound for the Lasso Created 2026-10-05 Updated 2026-10-06
For , the Lasso objective has this prediction bound whenever . Compare its value at and , apply Holder inequality to the score, and use the triangle inequality. No restricted eigenvalue condition is required.
Sobolev interpolation inequality 2026-10-14
For and , the displayed estimate holds on Euclidean space and a torus with the Fourier definition of the Sobolev norm. In the squared norm, write the weighted squared Fourier coefficient as the product of its weight- value to the power and its weight- value to the power . Holder inequality then proves the estimate. On a compact manifold, a finite coordinate cover and a partition of unity give the same estimate up to a constant. Young inequality implies the useful absorption form for nonnegative .
Strongly lacunary polynomial L1-L2 bound 2026-10-31
Use normalized measure on the circle. If a trigonometric polynomial has frequency support with , every zero sum of four signed frequencies pairs into opposite equal frequencies: the largest unpaired term would exceed the sum of the other three. Expanding the fourth moment and bounding its three possible pairings gives . The Holder inequality gives and hence the displayed estimate. Therefore an function with only these lacunary frequencies has square-summable coefficients: apply the estimate to its uniformly -bounded Fejér sums and then pass to finite coefficient blocks. Finitely many extra modes do not affect the conclusion.
Terminal scaling inequality for stochastic exponentials Created 2026-10-06 Updated 2026-10-14
For a convergent continuous local martingale and , factor into and an ordinary exponential. Apply the Holder inequality and then the concave Jensen inequality to the fractional power . The displayed lower bound is informative when the terminal exponential moment is finite.
Vinogradov mean-value method for a bilinear exponential sum Created 2026-10-06 Updated 2026-11-05
Expand moments of the inner sum and group equal power-sum differences. Their multiplicities are bounded by through the Cauchy-Schwarz inequality. Two applications of the Holder inequality yield a bound for containing and a product of short geometric-sum bounds over the moment differences. Thus a sharp mean-value estimate combines with rational approximation or spacing of the coefficients to prove cancellation. When all coefficients are integers, , so mean-value estimates alone cannot force cancellation.
Weak-strong product convergence lemma Created 2026-09-28 Updated 2026-10-14
If in , in , and , thenIndeed, split the difference into the weak pairing with and a term bounded by the Holder inequality using .