LBFGS for use in large maximum likelihood problem

Hi Robert, and others who are interested.

The following code gives a much simpler example of the use of lbfgs in a 
maximum likelihood problem where a symbolic sum noun does not suffice, 
but an explicit sum does...


-----------

kill(all);
load(distrib);
load(lbfgs);

data_test:random_normal(10,1,100);

pdf:exp(-(x-m)^2/(2*s^2))/(s*sqrt(2*%pi));


loglikelihood:'lsum(log(subst('x=x,pdf)),x,data_test);

lbfgs(-loglikelihood,[m,s],[9,2],.00025,[1,0]); /* ** THIS FAILS ** */

/* the error above is something like "non variable 2nd argument to diff 
9.0" */


explicit_expr:ev(-loglikelihood,nouns,infeval);

lbfgs(explicit_expr,[m,s],[9,2],.00025,[1,0]); /* ** THIS WORKS ** */