Technical Reports

2000

@techreport{Bas2000,
	vgclass =	{report},
	author =	{Sumit Basu},
	title =	{{ICA}: A Critical Review of Three Prominent Approaches},
	institution =	{Perceptual Computing Section, The MIT Media Laboratory},
	address =	{20 Ames St., Cambridge, MA 02139 USA},
	month =	{April~25},
	year =	{2000},
	url =	{http://web.media.mit.edu/~sbasu/papers/ICA_TR.pdf},
	abstract =	{We present the problem of Independent Components Analysis
	(ICA) and review three major approaches as described by Comon [9], Amari
	et al. [2], and Bell and Sejnowski [5]. We cast all three in a common
	notational framework, point out their strengths and weaknesses, and
	show how they are related to each other. We then go on to suggest
	several empirical studies to further investigate the numerical behavior
	of the algorithms. Finally, we present a number of novel
	extensions/applications of ICA. The most interesting of these is our
	notion of ``conditionally independent components analysis,'' in which we
	propose factoring the conditional density $p(x|y)$ to greatly reduce the
	amount of data needed to accurately model a scaled version of $p(y|x)$.},
}