Homotopy probability theory. II (Q496295): Difference between revisions

This is a companion paper of \textit{G. C. Drummond-Cole} et al. [J. Homotopy Relat. Struct. 10, No. 3, 425--435 (2015; Zbl 1401.55015)], where the authors present a deformation theoretic framework to explain and broaden a link between homotopy algebra and probability theory. Their expectation value is a linear map between a vector space \(V\) of random variables and the complex numbers that does not respect the product structure on \(V\). The failure of structure in the expectation value can be processed to give an infinite sequence of operations \(k_n :\) \(V^{ \otimes n} \to {\mathbb C}\). This sequence of operations assembles into an \(A_{\infty}\) morphism between two trivial \(A_{\infty}\) algebras. The main result in their first paper is that the \(A_{\infty}\) morphism ( \(L_{\infty}\) morphism in the commutative context) obtained via this process coincides with the cumulants of the initial probability space. In other words, Homotopy probability theory is concerned with the expected values of a chain complex of random variables equipped with an associative binary product. The product of random variables is assumed to be commutative throughout the paper. Let \(( C, e, a)\) be a commutative homotopy probability space, where \(C = ( V, d)\) is a chain complex over \({\mathbb C}\), \(e\) is a chain map \(C \to ({\mathbb C}, 0)\) (with \({\mathbb C}\) concentrated in degree 0), and \(a\) is an associative product \(S^2V \to V\), with the graded symmetric algebr \(SV\). An \(L_{\infty}\) algebra is a pair \((V, D)\) where \(V\) is a graded vector space and \(D\) : \(SV \to SV\) is a degree one coderivative satisfying \(D^2 = 0\). When \((V, D)\) is an \(L_{\infty}\) algebra with components \(d_n\) and \(A = ( W, d)\) is a commutative differential graded algebra, then \(( V \otimes W)\) can be given a canonical \(L_{\infty}\) algebra structure \(( V_A, D_A)\) whose first component is \(d_1 \otimes id\) \(+\) \(id \otimes d\) and whose \(n\)-th component for \(n > 1\) are given by \[ ( v_1 \otimes w_1, \dots , v_n \otimes w_n ) \mapsto d_n( v_1, \dots , v_n ) \otimes w_1 \cdots w_n. \] In particular, if \(A\) \(=\) \(( {\mathbb C}, 0)\) with the usual multiplication, then \(( V_A, D_A)\) is canonically isomorphic to \((V, D)\). When \(\Omega\) denotes the commutative differential graded algebra \({\mathbb C}[t, dt]\), polynomials in a variable \(t\) of degree 0 and its differential \(dt\), then for two \(L_{\infty}\) algebras \((V, D)\) and \((V', D')\), an \(L_{\infty}\) homotopy \(H\) from \((V, D)\) to \(( V', D')\) is an \(L_{\infty}\) morphism from \((V, D)\) to \(( V'_{\Omega}, D'_{ \Omega})\). Under these circumstances, moments and cumulants of a commutative homotopy probability space are given to be homotopy invariant. As a matter of fact, a collection of homotopy random variables \(( X_1, \dots , X_n)\) in a commutative homotopy probability space \((C, e,a)\) is defined to be the homotopy class of an \(L_{\infty}\) map \(( {\mathbb C}^n, 0)\) \(\to\) \(( V, D^a)\). While, the joint moment \(M(X_1, \dots, X_n)\) of \((X_1, \dots, X_n)\) is defined to be the \(L_{\infty}\) morphism \(M( X_1, \dots , X_n)\) \(=\) \(M \circ f\) where \(f\) : \(({\mathbb C}^n, 0)\) \(\to\) \((V, D^a)\) is any representative of \(( X_1, \dots , X_n)\) and \(M\) is the total moment of \(( C, e,a)\). The \(r\)-th joint moment of a collection of random variables is the \(r\)-th component of the joint moment \(M( X_1, \dots , X_n)\) and is a map \(m_r( X_1, \dots , X_n)\) : \(S^r{\mathbb C}^n \to {\mathbb C}\). Moreover, the joint cumulant \(C( X_1, \dots , X_n)\) of \(( X_1, \dots , X_n)\) is defined to be the \(L_{\infty}\) morphism \(K( X_1, \dots , X_n)\) \(=\) \(K \circ f\) where \(f\) : \(( {\mathbb C}^n, 0)\) \(\to\) \(( V, D^a)\) is any representative of \(( X_1, \dots , X_n)\) and \(K\) is the total cumulant of \((C, e,a)\). The \(r\)-th joint cumulant of a collection of random variables is the \(r\)-th component of \(C( X_1, \dots , X_n)\) and is a map \(k_r( X_1, \dots , X_n)\) : \(S^r {\mathbb C}^n \to {\mathbb C}\). Lastly the authors discuss the Gaussian case in one variable as well, that illustrates a toy example of how one might extend an algebra of random variables to a chain complex of random variables.