Appeal 2007-1088
Application 10/006,959
machine and that which has been predicted by a model-based machine.
Actual measurements or tests are discussed in paragraph 60 at page 4 of
Jelley. Moreover, the discussion from paragraph 61 to the discussion of
figure 11 at pages 4 and 5 of Jelley can clearly convey to the artisan the
comparative nature of neural network machines in a learning environment.
In making reference to figure 11 in paragraph 63, it is noted that substantial
agreement is shown between the actual measurements in a test environment
with those in a modelled environment. To the extent neural networks
inherently are learning devices, the implied need to be updated, if there were
not substantial agreement, is clearly conveyed to the artisan here. The
refinement necessary in the neural network environment of paragraph 18 in
the Summary of the Invention is conveyed as an updating-type function.
As noted by the Examiner at the bottom of page 13 of the Answer,
Appellants rely for the patentability of dependent claims 2 through 5, 8 and
9 upon the arguments presented with respect to independent claims 1 and 7
on appeal which we have found unpersuasive.
Turning to the rejection of independent claim 10 under § 103 relying
upon Jelley in view Talbott, we note the Examiner’s position reflects the
view that Jelley does not expressly disclose the feature of determining an
aging factor. This claimed feature is based on paragraph 40 of the
Specification at page 8, which explains that the number of hours of operation
of each machine may be logged and converted into an aging factor to
account for normal wear and tear of the machine. Jelley’s page 1,
paragraph 15, relied upon by the Examiner, acknowledges this aging
problem: “The existing [design] methods . . . generally assume that the wear
6
Page: Previous 1 2 3 4 5 6 7 8 Next
Last modified: September 9, 2013