Appeal 2007-1088 Application 10/006,959 development machine to the test machine” in claim 7 is broad enough to read on the same machine and model at different times. We turn now to the Appellants’ argument that the claimed comparing and updating features of the claims on appeal are not taught in Jelley. Appellants do not deny that Jelley discloses comparing data and using the results of the comparison to update the neural network. Instead, they deny that Jelly compares “data” from a test machine with “corresponding data” from the model development machine (claim 1) (Br. 14) or that he compares a “computed parameter” from a test machine to an “estimated parameter” from the test machine (claim 7) (Br. 16). These claim limitations also read on Jelley when the claimed machines and models are read on Jelley’s single machine and neural network at different times. Jelley compares a measured (i.e., actual) value of an operating parameter (e.g., lateral acceleration) to the value predicted for that operating parameter by the neural network (Jelley para. 61). Using the terms of claim 1, the predicted value corresponds to the recited “corresponding data of the model development machine” and the measured value corresponds to the recited “data from the at least one test machine.” In claim 7, on the other hand, both of the compared values are from the test machine and consist of (1) a “computed parameter” determined on the test machine and (2) an “estimated parameter” provided by the neural network on the test machine. The “computed parameter” corresponds to Jelly’s measured parameter value and the “estimated parameter” reads on Jelley’s predicted parameter value. Figure 11 of Jelley would clearly convey impliedly to an artisan a comparative functionality between a measurement obtained from a test 5Page: Previous 1 2 3 4 5 6 7 8 Next
Last modified: September 9, 2013