Took the docstrings from numpy.

theano/sandbox/linalg/ops.py

@@ -478,10 +478,12 @@ class CholeskyGrad(Op):
 class MatrixPinv(Op):
     """Computes the pseudo-inverse of a matrix :math:`A`.
 
-    Given a matrix :math:`A` of shape ``(d1, d2)``, ``pinv`` returns a
-    matrix :math:`A_{pinv}` of ``(d2, d1)`` such that the dot product
-    :math:`A_{inv} \cdot A` is as close as poosible to the identity
-    matrix :math:`I`.
+    The pseudo-inverse of a matrix A, denoted :math:`A^+`, is
+    defined as: "the matrix that 'solves' [the least-squares problem]
+    :math:`Ax = b`," i.e., if :math:`\bar{x}` is said solution, then
+    :math:`A^+` is that matrix such that :math:`\bar{x} = A^+b`.
+
+    Note that :math:`Ax=AA^+b`, so :math:`AA^+` is close to the identity matrix.
     """
     def __init__(self):
         pass