\magnification=1200
\baselineskip=20pt
\nopagenumbers
\font\big=cmr12 scaled \magstep2

\centerline{\bf STANFORD UNIVERSITY}
\centerline{\bf DEPARTMENT OF STATISTICS}
\centerline{\bf STANFORD / BERKELEY STATISTICS COLLOQUIUM}
\bigskip

\baselineskip=12pt
\centerline{4:10 p.m., Tuesday, November 5, 2002}
\centerline{Evans Hall, Room 60}
\centerline{Cookies:   3:30 -- Evans Hall, Room 1011}
\centerline{Reception: 5:10 -- Evans Hall, Room 1011}

\bigskip
\baselineskip=15pt
\centerline{\sl Trevor Hastie}
\centerline{\sl Statistics Department}
\centerline{\sl Stanford University}
\bigskip
\centerline{\bf Independent Component Analysis by Product Density Estimation}
\bigskip

The ICA model is a generalization of factor analysis, which gains
considerable power by avoiding Gaussian assumptions.  We present a
simple and direct approach for solving the ICA problem, using
density estimation and maximum likelihood.  Given a candidate
orthogonal frame, we model each of the coordinates using a
semi-parametric density estimate based on cubic splines. Since our
estimates have two continuous derivatives, we can easily run a
second order search for the frame parameters. Our method performs
very favorably when compared to state-of-the-art techniques, and has
a nice interpretation in terms of negentropy and mutual information.

* joint work with Rob Tibshirani

\bye