Revisions of the Guidelines for Patent Examination （2023）- the Analysis for the Invention Patent Application involving Computer Programs in Chapter 9 of Part II

View: 114     Date: 2024-01-17 08:06

Author: Ying ZHU

The Guidelines for Patent Examination （2023） was released recently, which makes adjustments to the rules of examination from multiple perspectives, such as improving the protection system for new form of industry and new fields, benchmarking international rules and protecting innovation achievements. Among them, the invention patent application involving computer programs in chapter 9 of part II, also has some significant adjustments, mainly involving the following four aspects:

Point 1 of the revision: involving the protection of computer program products invention

1.      The history review of the examination for "computer program products" in the Guidelines:

1985: Only considering computer hardware improvements, directly excluding the possibility for granting patent right for computer software inventions.

1993: Considering the combination of computer software and hardware, recognizing the auxiliary and subordinate status of software.

2001: For the first time to distinguish a computer program itself and an invention involving a computer program, the subject of protection is "method", and the technical features of the method are the executive steps of the invention involving a computer program.

2006: The protection subjects added "virtual device", that is, the device does not have to contain changes to the computer hardware, and adopted the "function module" drafting method corresponding to the "method" protection subject.

2017: The protection subjects added "computer hardware + software program" of "physical device", and added "readable storage media" protection subject.

2.      This revision clarifies that inventions involving computer programs can "protect computer programs or computer program products".

After allowing the storage media protection subject in 2017, the protection projects are further extended to the "program product" itself, and the way of drafting the claims can be "a computer program product/program instruction" and other ways to further meet the needs of the innovative entities, making the invention protection subjects involving computer programs more comprehensive coverage, and further benchmarking Europe, the United States, Japan, South Korea and other examination standards.

Point 2 of the revision: the examination benchmark involving artificial intelligence, big data and other technical fields "object".

For artificial intelligence/big data algorithm, there is no need to limit the data is a clear technical meaning in the technical field, if the internal relationship between the data conforms to the law of nature, it is in line with the object of patent protection.

      1. Examination logic involving object:

First, conducting the examination of Article 25, paragraph 1, paragraph 2 of the Patent Law, and then conducting the examination of article 2, paragraph 2 of the Patent Law.

For the examination of Article 25, paragraph 1, paragraph 2 of the Patent Law: if the claims only involve abstract algorithms or simple business rules and does not contain any technical features. It belongs to the rules and methods of intellectual activities stipulated in Article 25, paragraph 1 (2) of the Patent Law, and should not be granted patent rights. For example, the method of building a mathematical model without any technical features; And a rebate method that contains only the consumption amount of business rules.

If in addition to algorithmic features or business features, there are also technical features. Then the algorithm features or business features cannot be separated, should be based on the overall consideration of the technical scheme principle, can not only in the patent Law article 25, paragraph 1 (2) to exclude the possibility of obtaining a patent.

For the examination of Article 2, paragraph 2 of the Patent Law: it is also necessary to start from the overall consideration principle of the technical scheme, to judge whether the use of technical method in line with the law of nature has solved the technical problem, and to obtain the technical effect in line with the law of nature. That is, the three-element judgment method.

      2. Whether the internal relationship between the data conforms to the law of nature:

Starting from the three elements, although artificial intelligence and big data algorithms themselves have not improved the computer hardware, there is a specific technical correlation between the improvement of algorithms and the internal structure of the computer system, such as: algorithm setting or adjustment for the scheduling of hardware resources of the computer internal system, therefore, there is a technical correlation with the internal structure of the computer system. The technical purpose is to reduce the amount of data storage, reduce the amount of data transmission, improve the hardware processing speed and other technical problems, so as to obtain the technical effect of improving the internal performance of the computer system in line with the laws of nature, such as the improvement of hardware execution/processing efficiency and the improvement of accuracy. Therefore, it conforms to the provisions of Article 2, paragraph 2 of the Patent Law.

1) Analysis for examples in accordance with the laws of nature in the newly added examples of the Guidelines:

Analysis for example "Training Method of Deep Neural Network Model": according to the amount of training data can be adapted to single processor or multi-processor, then the model training algorithm and the internal structure of the computer system have a specific technical correlation, which can improve the performance of the processor to execute the training speed. Therefore, although the training data is not limited to the specific data in the specific application field, the correlation relationship between the training data and the internal structure of the computer system conforms to the law of nature, and therefore meets the object of patent protection.

Analysis for example "E-Coupon Usage Tendency": classify e-coupon types and train the identification model of e-coupon usage tendency according to the data of user behavior features, so as to predict the usage probability of e-coupon through the trained model. An internal correlation between the user behavior features and the tendency to use e-coupon. The correlation does not depend on the individual, but the behavior features of the consumer group, so it is in line with the law of nature.

2) Analysis of examples that do not conform to the law of nature in the newly added examples of the Guidelines:

Analysis for example "Price Prediction Method for Financial Products": the price of financial products follows the law of economics, and the law of economics is the behavior of the market. Therefore, the price prediction model does not determine the trend of future price, so it does not conform to the law of nature.

The adjustment of the object is also a very significant revision in the Guidelines, the revision is based on the basic examination logic of the object, in the examination process of Article 2, paragraph 2 of the patent Law, in the process of judging the three elements including technical problem, technical method and technical effect, which no longer limit whether the algorithm has a specific application field, no longer limit whether the data involved in the algorithm is the data with clear meaning in the specific technical field, but pay more attention to the internal correlation between the data in accordance with the law of nature, not only solves the confusion for innovative entities about the limited "model" application circumstances, but also specifies the model algorithm that does not conform to the law of nature. For example, the algorithms that follow the law of economics, sociology and other unnatural laws are excluded.

Point 3 of the revision: inventiveness examination involving the algorithm to achieve internal performance improvement of the system.

When the algorithm achieves internal system performance improvement, the contribution of the algorithm to the technology should be considered.

In the examination involving the subject matter of Article 2 (2) of the Patent Law, " there is a specific technical correlation between the improvement of algorithms and the internal structure of the computer system, such as: algorithm setting or adjustment for the scheduling of hardware resources of the computer internal system, therefore, there is a technical correlation with the internal structure of the computer system. The technical purpose is to reduce the amount of data storage, reduce the amount of data transmission, improve the hardware processing speed and other technical problems, so as to obtain the technical effect of improving the internal performance of the computer system in line with the laws of nature, such as the improvement of hardware execution/processing efficiency and the improvement of accuracy." The examination logic is also applicable to inventiveness examination, it can be considered that the algorithm features and technical features in the function of each other support, there is an interaction relationship, should consider the contribution of the algorithm.

Analysis for example "A method for adapting neural network parameters": for different neural network architectures, adapting different hardware resources, the incoming parameters of each layer of the neural network, the determination of parameter dimensions, and the size on each dimension can be determined based on the utilization rate of the hardware computed by the neural network. Therefore, the parameter adaptation of neural network improves the operational effect of the hardware, that is to say, the algorithm features and technical features of the neural network parameters are functionally supported by each other, and there is an interaction relationship. The problem to be solved is how to make the hardware efficiently perform the operations in the neural network, so it is necessary to consider the technical contribution.

The contribution of algorithm features in the inventiveness evaluation has always been the focus of the industry discussion, and this revision also gives a clear idea for examination, on the basis of the overall consideration principle of not separating algorithm features or commercial features, pay attention to whether the algorithm features achieve the improvement of the internal performance of the computer system, or, whether the combination of algorithmic features and technical features has improved the internal performance of the computer system, and accordingly consider the contribution of algorithmic features. This examination idea also provides clear direction for the subsequent mining and improvement of relevant technologies, and how to proceed with the invention application.

Point 4 of the revision: the consideration of user experience improvement in inventiveness evaluation.

The improvement of user experience is directly brought or generated by technical features, or is jointly brought or generated by technical features and algorithmic features or business rules and method features that support and interact with each other functionally, and should be taken into account in the inventiveness examination.

1.      What is user experience

Sensory experience, psychological experience, emotional experience, etc. are all user experience, such as: operation convenience, interface visual sensory experience, etc., program response speed and so on.

2.      User experience is not individual experience, the problem of user experience is technical problem essentially

Analysis for example added in the Guidelines "A method of logistics distribution": to solve the problem of how to effectively improve the efficiency of goods distribution and reduce distribution costs, through the server according to the hand-held terminal ID, current location and distribution scope, to push order information in batches and other technical method to achieve the technical effect of improving distribution efficiency and reducing distribution costs. That is, it improves the experience of the dispatcher and also improves the user experience of ordering. The improved experience is brought about by both technical features and business features related to distribution, so it should be taken into account when evaluating inventiveness. On the contrary, if the improvement in delivery efficiency is only brought about by the increase in the amount of red envelope by the user/platform, then the improvement in user experience is only brought about by the business features, then it should not be taken into account when evaluating inventiveness.

Artificial intelligence, big data related fields of technology, the starting design point of the product is often from the perspective of user experience, this revision makes it clear that the user experience is not individual experience, and further explore its essence, sensory/psychological/emotional experience is poor, its essence may be caused by inconvenient operation, interface display effect is not smooth, program response speed is slow and other technical problems. Therefore, the essence of the user experience problem is technical problem, and the improvement of user experience is directly brought or generated by technical features, or is jointly brought or generated by technical features and algorithmic features or business rules and method features that support and interact with each other functionally, and should be taken into account in the inventiveness examination.

The revised Guidelines for the invention patent application involving computer programs in chapter 9 of part II, covering the protection of the subject, object, inventiveness examination and other aspects, to further meet the protection needs of innovative entities in new form of industry and new fields, and is of great significance for stimulating innovation.