Appeal No. 95-0633
Application 07/728,426
The invention pertains to a method for training a neural
network classifier.
Representative claim 22 is reproduced as follows:
22. A method of training a neural network classifier,
comprising the steps of:
(a) providing a first set of target points Z , Z , ... Z1 2 L
in a feature space;
(b) forming an estimated target probability density P on
said feature space from said target points Z , Z , ... Z ;1 2 L
(c) providing a second set of target points W , W , ... W1 2 M
in said feature space;
(d) defining a threshold T from the number of W with
j
P(W ) > T and the number of W with P(W ) < T;j j j
(e) providing a third set of points X , X , ... X in said1 2 N
feature space, and forming a set of pairs (X , Y ) where Y isj j j
“target” when P(X ) > T and Y is “clutter” when P(X ) < T; andj j j
(f) using the pairs (X , Y ), (X , Y ), ..., (X , Y ), ...,1 1 2 2 j j
(X , Y ) as input/output pairs to train a neural networkN N
classifier.
The examiner relies on no references.
Claims 22-25 stand rejected under 35 U.S.C. § 101 as
being directed to nonstatutory subject matter in the form of a
mathematical algorithm. Claims 22-25 also stand rejected
under 35 U.S.C. § 112, second paragraph, as failing to
particularly point out and distinctly claim the invention.
2
Page: Previous 1 2 3 4 5 6 7 8 Next
Last modified: November 3, 2007