• In this paper, we study the partial bi-free $S$-transform of a pair $(a,b)$ of random variables, and the $S$-transform of the $2\times 2$ matrix-valued random variable $\left(\begin{matrix}a&0\\0&b\end{matrix}\right)$ associated with $(a,b)$ when restricted to upper triangular $2\times 2$ matrices. We first derive an explicit expression of bi-free multiplicative convolution (of probability measures on the bi-unit-sphere $\mathbb{T}^2$ of $\mathbb{C}^2$, or on $\mathbb{R}^2_+$ in $\mathbb{C}^2$) from a subordination equation for bi-free multiplicative convolution. We then show that, when $(a_1, b_1)$ and $(a_2,b_2)$ are bi-free, the $S$-transforms of $X_1=\left(\begin{matrix}a_1&0\\0&b_1\end{matrix}\right)$, $X_2=\left(\begin{matrix}a_2&0\\0&b_2\end{matrix}\right)$ satisfy Dykema's twisted multiplicative equation for free operator-valued random variables if and only if at least one of the two partial bi-free $S$-transforms of the pairs of random variables is the constant function 1 in a neighborhood of $(0,0)$. This is the case if and only if one of the two pairs, say $(a_1,b_1)$, has factoring two-band moments (that is, $\varphi(a_1^mb_1^n)=\varphi(a_1^m)\varphi(b_1^n)$, for all $m,n=1, 2, \cdots$). We thus find tons of bi-free pairs of random variables to which the $S$-transforms of the corresponding matrix-value random variables do not satisfy Dykema's twisted multiplicative formula. Finally, if both $(a_1,b_1)$ and $(a_2,b_2)$ have factoring two-band moments, we prove that the $\Psi$-transforms of $X_1$, $X_2$, and $X_1X_2$ satisfy a subordination equation.
  • In this paper, we prove four-moment theorems for multidimensional free Poisson limits on free Wigner chaos or the free Poisson algebra. We prove that, under mild technical conditions, a bi-indexed sequence of free stochastic integrals in free Wigner algebra or free Poisson algebra converges to a free sequence of free Poisson random variables if and only if the moments with order not greater than four of the sequence converge to the corresponding moments of the limit sequence of random variables. Similar four-moment theorems hold when the limit sequence is not free, but has a multidimensional free Poisson distribution with parameters $\lambda>0$ and $\alpha=\{\alpha_i: 0\ne \alpha_i\in \mathbb{R}, i=1, 2, \cdots\}$.
  • Inspired by R. Speicher's multidimensional free central limit theorem and semicircle families, we prove an infinite dimensional compound Poisson limit theorem in free probability, and define infinite dimensional compound free Poisson distributions in a non-commutative probability space. Infinite dimensional free infinitely divisible distributions are defined and characterized in terms of its free cumulants. It is proved that for a distribution of a sequence of random variables, the following statements are equivalent. (1) The distribution is multidimensional free infinitely divisible. (2) The distribution is the limit distribution of triangular trays of families of random variables. (3) The distribution is the distribution of $\{a_1^{(i)}: i=1, 2, \cdots\}$ of a multidimensional free Levy process $\{\{a_t^{(i)}:i=1, 2, \cdots\}: t\ge 0\}$. (4) The distribution is the limit distribution of a sequence of multidimensional compound free Poisson distributions.
  • We study two-faced families of random variables having bi-free infinitely divisible distributions. We prove a limit theorem of the sums of bi-free two-faced pairs of random variables within a triangular array. Then, by using the full Fock space operator model, we show that a two-faced pair of random variables has a bi-free (additive) infinitely divisible distribution if and only if its distribution is the limit distribution in our limit theorem. Finally, we characterize the bi-free (additive) infinite divisibility of the distribution of a two-faced pair of random variables in terms of bi-free Levy processes.
  • We prove a multidimensional Poisson limit theorem in free probability, and define joint free Poisson distributions in a non-commutative probability space. We define (compound) free Poisson process explicitly, similar to the definitions of (compound) Poisson processes in classical probability. We proved that the sum of finitely many freely independent compound free Poisson processes is a compound free Poisson processes. We give a step by step procedure for constructing a (compound) free Poisson process. A Karhunen-Loeve expansion theorem for centered free Poisson processes is proved. We generalize free Poisson processes to a notion of free Poisson random measures (which is slightly different from the previously defined ones in free probability, but more like an analogue of classical Poisson random measures). Then we develop the integration theory of real-valued functions with respect to a free Poisson random measure, generalizing the classical integration theory to the free probability case. We find that the integral of a function (in certain spaces of functions) with respect to a free Poisson random measure has a compound free Poisson distribution. For centered free Poisson random measures, we can get a simpler and more beautiful integration theory.
  • A triple of finite von Neumann algebras $B\subseteq N\subseteq M$ is said to have the relative weak asymptotic homomorphism property if there exists a net of unitary operators $\{u_{\lambda}\}_{\lambda\in \Lambda}$ in $B$ such that $$\lim_{\lambda}|\mathbb{E}}_B(xu_{\lambda}y)-{\mathbb{E}}_B({\mathbb{E}}_N(x)u_{\lambda}{\mathbb{E}}_N(y))\|_2=0$$ for all $x,y\in M$. We prove that a triple of finite von Neumann algebras $B\subseteq N\subseteq M$ has the relative weak asymptotic homomorphism property if and only if $N$ contains the set of all $x\in M$ such that $Bx\subseteq \sum_{i=1}^n x_iB$ for a finite number of elements $x_1,...,x_n$ in $M$. Such an $x$ is called a one sided quasi-normalizer of $B$, and the von Neumann algebra generated by all one sided quasi-normalizers of $B$ is called the one sided quasi-normalizer algebra of $B$. We characterize one sided quasi-normalizer algebras for inclusions of group von Neumann algebras and use this to show that one sided quasi-normalizer algebras and quasi-normalizer algebras are not equal in general. We also give some applications to inclusions $L(H)\subseteq L(G)$ arising from containments of groups. For example, when $L(H)$ is a masa we determine the unitary normalizer algebra as the von Neumann algebra generated by the normalizers of $H$ in $G$.