EP 3 690 761 A1 relates to the provision of adaptive deep learning models calibrated and personalized for “users” of autonomous vehicles.
Brief outline of the case
The application was refused as the MR, AR2 and 3 lacked IS over D1= US2018/0053102 A1.
AR1 filed during OP was not admitted.
The board dismissed the appeal.
The applicant’s point of view
The applicant submitted inter alia, that to arrive at the invention starting from D1 a number of modifications were needed. There was no reason for the skilled person to perform all of them. The added features, in particular in the AR4, further increased the already large number of differences over D1.
In its reply to the board’s communication under Art 15(1), and during the OP, the applicant discussed the conceptual differences between D1 and claim 1, but did not contest the set of differences between claim 1 and D1 as identified by the board.
The applicant argued that the ED and the Board’s analysis was ex post facto. Without knowledge of the invention the skilled person had no reason to modify D1 in a way so as to arrive at the claimed invention.
The invention related to continuous learning of a deep learning model for a specific autonomous vehicle. The model was retrained with specific video data for specific circumstances and stored in a library containing the various models.
The storage of models retrained for various circumstances allowed for efficient fine-tuning through fast re-training with minimal data. The selection step based on video data information ensured that the proper model was selected and updated.
In contrast, the focus of D1 was to develop a real-time solution suitable for onboard use. The solution of D1 was one in which a single generic stock model was adapted to a driver on the vehicle itself, during the operation of the vehicle.
For this reason, there was no need for a library in D1. The ED merely stated that the stock model had to be stored somewhere and concluded that this already disclosed a library.
The skilled person also had no reason to perform the model adaptation on a server. There was enough computing power on a vehicle to perform re-training, and the need to communicate with a server might compromise real-time adaptation.
In fact, the real-time requirement of D1 taught away from a centralized solution. Sending video data, waiting for computation and receiving the adapted model caused time delays which did not allow real-time adaptation.
The board’s decision
In its communication under Art 15(1) RPBA, the board had identified a set of four differences between claim 1 and D1.
The board held that all those differences were obvious and could not justify IS.
The board also noted that the applicant’s conceptual presentation of the invention did not entirely correspond to the claimed invention, which is less detailed and therefore of broader scope.
The board observed that for IS, the question is not what the CPA discloses, but how the skilled person would modify it, e.g. in order to improve it. The board added that, in general, the skilled person would consider well-known alternatives.
The applicant did not dispute that information was missing from the claims, and even from the application as a whole, but argued that the skilled person skilled would know how to implement the claimed steps so as to obtain the desired technical effect.
This was not convincing for the board, which concluded that the skilled person, knowing how to improve accuracy by cross-validation between automatic and manual labelling, would use cross-validation without exercising any inventive skill.
The board observed that the number of differences over a certain piece of prior art is neither decisive nor a reliable indicator for the presence of IS.
First, the number of differences itself may be, and often is, deceiving. One modification may imply or make obvious several other differences. Also, in complex systems it is very easy to accumulate a large number of individual differences while simply considering the different options available to the person skilled in the art.
Secondly, whether several modifications combine to provide an inventive overall contribution does not depend on their number. For instance they may be obvious solutions to independent, “partial problems“.
Ultimately, the claimed invention must contain a new and non-obvious technical teaching. The board did not see such a non-obvious teaching reflected in any of the requests on file.
Comments
The board is to be followed when it states that a large number of differences do not necessarily imply IS.
The board need not be followed when it claims that IS goes on a par with an improvement over the CPA. When the OTP boils down to finding an alternative solution, where is the improvement?
When discussing the IS of a given claim, what matters are claimed features, and not features which might be either in dependent claims or in the description, or even not disclosed at all.
It might be sometimes necessary to rely on a large number of pieces of prior art in order to arrive to a claimed invention, but it can well be that one piece of prior art leads to the next one, or in the absence of synergy between all the differences, each difference might be obvious for itself.
Comments
Leave a comment