Appeal No. 2000-0866 Page 9
Application No. 08/852,829
Puskorius '750 does teach (column 5, lines 52-55) that trained
networks can be embedded into dedicated neural network
hardware chips, the combined teachings of the applied prior
art do not teach or suggest the claimed steps of calibrating a
simulator for simulating operation of the vehicle component
using the test data, generating at least one map which
characterizes performance of the vehicle component as a
function of predetermined parameters, the map being based on
output of the simulator for a second set of operating
conditions, and embedding the trained neural network into the
controller.
In our view, the only suggestion for modifying Puskorius
'700 in the manner proposed by the examiner to meet the above-
noted limitations stems from hindsight knowledge derived from
the appellants' own disclosure. The use of such hindsight
knowledge to support an obviousness rejection under 35 U.S.C.
§ 103 is, of course, impermissible. See, for example, W. L.
Gore and Assocs., Inc. v. Garlock, Inc., 721 F.2d 1540, 1553,
220 USPQ 303, 312-13 (Fed. Cir. 1983), cert. denied, 469 U.S.
851 (1984).
Page: Previous 1 2 3 4 5 6 7 8 9 10 11 12 13 Next
Last modified: November 3, 2007