Using a non-Gaussian prior
	• If the prior distributions on the factors are not Gaussian, some orientations will be better than others – It is better to generate the data from factor values that have high probability under the prior. –  one big value and one small value is more likely than two medium values that have the same sum of squares.
				If the prior for each hidden     activity is     the iso-probability contours are straight lines at 45 degrees.