---
title: Mixed Weak Compactness and CMO Property
url: https://www.emergentmind.com/topics/mixed-weak-compactness-cmo-property
type: topic
---

# Mixed Weak Compactness and CMO Property

The mixed weak compactness and CMO property encompasses a family of phenomena governing the weak compactness of operator classes and tensor products, particularly in Banach space theory, multilinear operator ideals, and bi-parameter Calderón-Zygmund theory. The product weak compactness property, known as property (AW) in Banach space literature, asserts that if the set of tensors built from subsets of two spaces is weakly precompact, at least one factor must be relatively norm compact. In the bi-parameter singular integral context, the mixed weak compactness/CMO property characterizes operators whose matrix coefficients vanish in mean oscillation as parameters recede to infinity. These properties interface with classical ideals (absolutely summing, strongly summing, factorable p-summing), underlie key factorization theorems, and are essential for operator compactness, $\ell_1$ embeddability, and paraproduct decompositions.

## 1. Foundational Properties and Definitions

Weak precompactness for a set $W$ in a Banach space $Z$ consists of every sequence in $W$ admitting a weakly Cauchy subsequence, equivalently forbidding $\ell_1$-sequences (Rosenthal's theorem) [2305.06089]. Relative norm compactness is stricter, requiring norm-closure compactness or that every sequence has a norm-convergent subsequence.

For Banach spaces $X$ and $Y$, the projective tensor product $X \widehat{\otimes}_\pi Y$ is the completion of their algebraic tensor product under the norm
$$
\|z\|_\pi = \inf\left\{ \sum_{i=1}^n \|x_i\|_X \|y_i\|_Y : z = \sum_{i=1}^n x_i \otimes y_i \right\}.
$$

**Product weak compactness property (AW)**: A pair $(X, Y)$ satisfies (AW) if for any sets $W_1 \subseteq X$, $W_2 \subseteq Y$, whenever
$$
\{ x \otimes y : x \in W_1, y \in W_2 \}
$$
is weakly precompact in $X \widehat{\otimes}_\pi Y$, then at least one of $W_1$ or $W_2$ is relatively norm compact [2305.06089]. This property mediates factorization in operator theory and is pivotal in the analysis of multiplication operators and tensor embeddings.

## 2. Sufficient Conditions and Structural Techniques

The principal sufficient condition for (AW) involves the “coarse $p$-limited” property, designated $(R_p)$. A Banach space $X$ has $(R_p)$ if every weakly compact yet non-norm-compact subset exhibits its failure via an operator $u : X \to \ell_p$ making $u(W)$ non-norm-compact. If $X$ has $(R_p)$ and $Y$ has $(R_q)$ with $1/p + 1/q > 1$, then $(X, Y)$ has (AW) [2305.06089]. This result harnesses classical sequence space techniques and is exemplified by $(\ell_p, \ell_q)$ for $1/p + 1/q \geq 1$.

In the context of strongly weakly compactly generated (SWCG) spaces [2206.08651], the existence of a single weakly compact generator set $G$ that strongly generates all weakly compact subsets ensures preservation of SWCG in projective tensor products, given unconditional finite-dimensional decompositions (FDDs) with disjoint lower $p$- and $q$-estimates and $1/p+1/q \geq 1$. These decompositions allow controlling “far-out” blocks and transferring weak compactness through tensor products.

## 3. Mixed Weak Compactness/CMO in Bi-Parameter Singular Integral Theory

In bi-parameter Calderón-Zygmund frameworks [2601.05454], the **product weak compactness property (PWC)** is quantified via wavelet matrix coefficients. Let $T$ be a bi-parameter singular integral operator and $\{\psi_z\}$ a continuous wavelet frame. PWC means that for every fixed radius $R>0$,
$$
\lim_{z \to \infty} \sup_{w_1 \in D(z_1, R),\ w_2 \in D(z_2, R)} |\langle T\psi_z, \psi_w \rangle| = 0,
$$
where $z = (z_1, z_2)$ and $D(z_i, R)$ are hyperbolic balls in the $ax+b$ group.

**Mixed weak compactness/CMO property** dictates, for fixed parameters, that
$$
\langle T(1 \otimes \psi_z), \cdot \otimes \psi_w \rangle \in \mathrm{CMO}(\mathbb{R}^{n_1}),
$$
with uniform vanishing of the $\mathrm{BMO}$ norm as $z \to \infty$, analogously for other parameter roles. The simultaneous satisfaction of PWC, mixed weak compactness/CMO, and vanishing paraproduct distributions in $\mathrm{CMO}$ ensures compactness of $T$ on $L^2(\mathbb{R}^{n_1} \times \mathbb{R}^{n_2})$. This interaction is central to the compact $T1$ theorem for bi-parameter SIOs. The PWC is both necessary and, in concert with the other hypotheses, sufficient for operator compactness, validated by a reduction to fully cancellative operators and invocation of the abstract localization criterion (Theorem 3.7).

## 4. Multilinear Operator Ideals and Factorization

The product weak compactness property extends to multilinear operator contexts as formulated in [1311.4685]. While absolutely $p$-summing linear operators are weakly compact, strongly $p$-summing multilinear operators (class $\mathbb{S}\mathrm{t},p$) do not, in general, inherit weak compactness, nor do they factor through $L^p$ in the classical sense.

A refined subclass, **factorable strongly $p$-summing operators (FS$\pi_p$)**, is defined via an $\ell_1$-$\ell_p$ mixed summability condition. For a map $T : X_1 \times \cdots \times X_n \to Y$, the defining estimate is
$$
\left( \sum_{j=1}^{m_1} \sum_{i=1}^{m_2} |\lambda_i^j| \|T(x_{1,i}^j, \ldots, x_{n,i}^j)\| \right)
\leq C \cdot \sup_{\varphi \in B(L(X_1, \ldots, X_n; \mathbb{K}))} \left( \sum_{j=1}^{m_1} \sum_{i=1}^{m_2} |\lambda_i^j|^p |\varphi(x_{1,i}^j, \ldots, x_{n,i}^j)|^p \right)^{1/p}.
$$
This subclass (FS$\pi_p$) admits a Pietsch-type factorization through $L^p(\mu)$-spaces and is automatically weakly compact. Thus, FS$\pi_p$ operators possess the product weak compactness property as multilinear extensions of the ideal $\Pi_p$ [1311.4685].

## 5. Applications and Concrete Examples

Applications span operator theory, tensor product structure, and $\ell_1$-embeddability.

- **Multiplication operators**: For $R, S \in \mathcal{L}(X)$, the map $P_{R,S}(T) = R T S$ is weakly compact iff $S \otimes R^* : X \otimes_\pi X^* \to X \otimes_\pi X^*$ is weakly compact. Property (AW) forces, under $(R_p)$, that either $R$ or $S$ is compact, recovering Saksman–Tylli theorems for $X \subseteq \ell_p$ or James space [2305.06089].

- **$\ell_1$-embedding and compact operators**: Absence of $\ell_1$-copies in $X \widehat{\otimes}_\pi Y$ is characterized by compactness of all operators $X \to Y^*$, under weak precompactness criteria [2305.06089].

- **Banach lattices and function spaces**: SWCG is preserved under tensor products for $L_p[0,1]$ and classical Banach lattices with unconditional FDDs and compatible $p$- and $q$-estimates ($1/p+1/q \geq 1$) [2206.08651].

A representative table relates space structures to AW and SWCG properties:

| Space Pair                   | Condition                           | Property Holds      |
|------------------------------|-------------------------------------|--------------------|
| $(\ell_p, \ell_q)$           | $1/p+1/q \geq 1$                    | AW, SWCG           |
| $(X, X^*)$ subspace of $\ell_p$ | X has unconditional FDD, $(R_p)$   | AW                 |
| (James space $J$, $J^*$)     | Both have $(R_2)$                   | AW                 |

## 6. Abstract Compactness and Criteria

The abstract compactness criterion in [2601.05454] synthesizes Hilbert space tensor structures and frame decompositions. For $T : H_1 \otimes H_2 \to H_1 \otimes H_2$, with matrix blocks $T_{(z,w)} : H_1 \to H_1$ formed via frames $\{\psi_z\}$ and partial localization, compactness of $T$ follows given:

- Each $T_{(z,w)}$ is compact on $H_1$.
- Norms $\|T_{(z,w)}\|$ vanish uniformly off the diagonal as $(z,w) \to \infty$.

Verification of these hypotheses in bi-parameter settings typically employs:

- **Kernel-integral criteria**: Decay of kernel coefficients under integration against frame elements.
- **Diagonal CMO and compact kernel criteria**: Oscillation vanishing on the diagonal suffices to imply the requisite uniform decay.

Thus, product weak compactness and its mixed versions serve as foundational constraints for operator compactness in both linear and multilinear contexts, guiding factorization, tensor product structure, and harmonic analysis applications. This framework identifies broad classes of Banach spaces and operator ideals governing compactness phenomena on both function spaces and operator algebras [2305.06089, 2206.08651, 1311.4685, 2601.05454].

Source: https://www.emergentmind.com/topics/mixed-weak-compactness-cmo-property