Patenting Machine Learning Tech at USPTO vs EPO

Artificial intelligence technology has been around for a long time, but has recently made advances prompting recognition as the transformative force that it truly is. While applicants have successfully patented artificial intelligence inventions for many years, the US has been more favorable than Europe for some types of AI. Here we focus on one area of difference between patent-eligibility of NLP inventions in the USPTO versus the EPO. We use board decisions for distinguishing the two jurisdictions. 

As some background, machine learning is the AI technique most frequently disclosed in patents, and is included in more than a third of all AI-related patent applications.

aiimage

Photo credit Aglaé Bassens et al., “Deep Learning Illustrated: A Visual, Interactive Guide to Artificial Intelligence” (August 5, 2019). 

Machine learning is an AI technique ever-growing in dominance. Deep learning and neural networks are fastest growing of the lot. This is at least partially represented in patent filings, where filings of machine learning patent applications have grown annually on average by 28% from 2013 to 2016. This is notably higher than the average annual growth across all new areas of technology, which was 10% during the same time. Within this category, deep learning, used for example in speech recognition, had a 175% average annual growth rate from 2013 to 2016. 

Among functional applications, computer vision is the most popular, and is mentioned in almost half of AI-related patents. The next hottest area in functional applications is natural language processing. Examples of NLP in industry include classifying documents; machine translation; search engine optimization; speech recognition; and chatbots. We focus here on patent-eligibility of NLP in this blog post. 

Source: WIPO Technology Trends 2019, “Artificial Intelligence,” at 14 and 31.

In the US, machine learning applications have generally fared well for patentability purposes. Even though many of these machine learning inventions are rooted in software, and presumably vulnerable to Alice-type eligibility rejections, allowance rates have been substantially higher than other software classes. For example, comparing class 706 (Artificial Intelligence: Data Processing) to class 705 (Financial, Business Practice, Management, or Cost/Price Determination: Data Processing), we see a big difference. 

allowanceai

https://developer.uspto.gov/visualization/allowance-rate-uspc-class

One reason why these allowance rates are much lower may depend on the art unit differences. Machine learning inventions typically get assigned to the 2121 or 2122 art units whereas business method inventions get assigned to the 3620s, 3680s and 3690s. The 3600 art units are well known for applying knee-jerk Section 101 patent-ineligible rejections whereas AI art units are not as preoccupied with Section 101. Often times, Examiners in these machine learning art units see the cutting edge technology of machine learning in these applications and generally quickly grant the patents for these inventions. 

But not all AI inventions are as easy to get allowed, especially depending on the jurisdiction. Take Europe, for example. The standard for patent-eligibility at the EPO is somewhat different than the US in that it requires a sufficiently technical nature (i.e., the claim must have a technical implementation or technical application). For image processing and speech recognition, this technical nature can be easily shown. But other types of machine learning tech, such as NLP, have not been so recognized as having a technical purpose.  

As pointed out in this Marks & Clerk blog post, there is a historical context to difficulties in patenting some NLP technologies. 

In T 52/85, the Board considered a system for automatically generating a list of expressions whose meaning was related to an input linguistic expression.  The Board held that the relationship between the input and output expressions was not of a technical nature, but was instead a matter of their “abstract linguistic information content”.  The Board consequently found that the claimed subject-matter was unpatentable.

In another relatively old decision, T 1177/97, the technology at issue related to machine translation.  The Board again found the claimed subject-matter to be unpatentable, stating “Features or aspects of the method which reflect only peculiarities of the field of linguistics, however, must be ignored in assessing inventive step.”  This statement is often quoted by examiners when applying the Comvik approach to inventions in the field of natural language processing. Although the Board in T 1177/97 also held that “information and methods related to linguistics may in principle assume technical character if they are used in a computer system and form part of a technical problem solution”, it is hard in practice to convince the EPO that a technical problem is solved by the linguistic aspects of an invention. 

https://www.marks-clerk.com/Home/Knowledge-News/Articles/Patenting-Artificial-Intelligence-at-the-European.aspx#.XauaEZNKjBI

The US has been more favorable about the patentability of NLP technology. For example, the PTAB has recently reversed patent-eligibility rejections in a large proportion of AI applications. Results will follow. For example, one PTAB panel recently overturned a NLP-related invention as passing the two-step Alice framework. In Ex parte BAUGHMAN et al., Appeal No. 2019-000665 (PTAB Sept. 25, 2019), the PTAB overturned an Examiner’s rejection for the claims being directed to an abstract idea. 

claim2.pngclaim1claim2

The Board, under step 2A prong 2, reasoned that the claim recites additional elements, which are outside the abstract idea, that include: “receiving, by the question answering system, a function call comprising an input question and a set of non-local context evidence in closure form.” The Board explained that the recited use of a “function call” and the use of “closure form” are particular (non-generic) software technology limitations. Specifically, the “function closure”-related software limitations recited in claim 1’s first step are integrated with the limitations that describe the abstract idea for generating answers to a question. 

The Board viewed the claim holistically by stating that “[t]aken as a whole, claim 1 recites a set of steps for a particular query- and hypothesis-based processing sequence and set of rules, executed by a QA system.” Then citing McRO, the Board held that this amounts to “us[ing] the limited rules in a process specifically designed to achieve an improved technological result in conventional industry practice,” i.e., to improve the technology of QA systems.” After coming to this determination, the Board found that the claim imposes meaningful limits on the application of the recited judicial exception for generating candidate answers to a question and thus are not directed to a patent-ineligible abstract idea. 

AI will continue to transform all sectors of industry and patentability standards across jurisdictions will continue to change. Patentability standards across jurisdictions should continue to evolve to balance the growing impact of AI on society. As it does so, it is important to anticipate prosecution strategy internationally with the best patent data. 

PTAB reverses abstract idea rejection of business method patent application – Some key practice tips

There has been much energy focused on the flawed nature of the current Alice/Mayo framework for evaluating patent-eligibility. Even the well-intentioned desires of USPTO director Andrei Iancu’s to move away from this framework has its limits.

In a recent decision, Ex Parte Pasupulati et al, the Board reversed an abstract idea rejection to a claim directed to integrating an API-based mechanism to a merchant’s website without the merchant having to understand the API or having to know how to code in a language specific to the API. The Board rebuffed the Examiner’s position for both steps.

Step 1

Under Step 1, the Board first found that the Examiner had improperly overgeneralized the claims as “payment management in a network” or the general concept of “coding”. Without being able to over simplify the claim, it makes it much more difficult for the claims to be “directed to” an abstract idea.

Next, the Board found that the focus of the claims is not on judicical exceptions (i.e., “do not recite a mathematical algorithm … [n]or do they recite a fundamental economic or longstanding commercial practice” identified as an “abstract idea” by the Supreme Court in its trilogy of Benson, Flook, and Diehr, and its recent duet of Bilski and Alice). Instead, the Board found that the claims provide a particular solution to the particular problem of integrating an API-based mechanism to a merchant’s website without the merchant having to understand the API or having to know how to code in a language specific to the API.

Still, the Board did not have the confidence to rely only on Step 1 in reversing this rejection because “identifying the precise nature of the abstract idea [of Appellants’ claims] is not as straightforward,” it proceeded to Step 2.

Step 2

The Board analogized to DDR Holdings in showing that the claims contain an “inventive concept” for two reasons. The first reason is somewhat related to the step 1 analysis. That is, the Board found that the claims provide a technical solution to a technical problem unique to the Internet. Here it was generating code snippets and files that can be automatically integrated with an online website so that an API-based mechanism can be used with the online website.

The second reason analogized to Bascom. The Board found that the combination of Appellants’ claim elements, like the claims in Bascom, provides a practical application of integrating an API based mechanism to a merchant’s website without the merchant having to understand the API or having to know how to code in a language specific to the APL This transforms any abstract idea (i.e., “coding” or “payment management in network”) into “a particular, practical application of that abstract idea.”

Conclusion

Despite the chorus of criticism of the Alice/Mayo framework continuing to grow, it is the law of the land. That being said, there are ways to navigate through its complexities. This decision nicely shows practice tips that can be used for similarly rejected applications. For step 1, one practice tip is not to let the examiner oversimplify what the claims are directed to. Consider focusing of the claims with its technical features as a problem to solve. For step 2, if applicable, consider showing how the claimed invention is solving a problem unique to a computer-specific environment (e.g., the Internet). Lastly, consider how the combination of claim elements might transform the alleged abstract idea into a particular, practical application.

Another Section 101 lack of utility rejection bites the dust

As reported in today’s recap email, in Ex Parte GELFAND et al (Aug. 2, 2018), the Board reversed a lack of utility rejection. This continues a longstanding trend of the Board frequently reversing such rejections.

In rejecting the claims as lacking utility under Section 101, the Examiner had asserted that the claims were not supported by a specific, substantial, and credible asserted utility, or a well-established utility. For support, the Examiner relied on MPEP § 2107.01 B(C), which states that “[a] method of assaying for or identifying a material that itself has no specific and/or substantial utility” is not considered to have a substantial utility.”

On appeal, the appellant argued “The present invention relates to the field of nucleic acid detection” and that nucleic acid detection has numerous uses including diagnostic uses and research uses–a specific utility.

While acknowledging the Examiner’s cited MPEP section, the Board rejected the Examiner’s reasoning. That identifying the presence of target nucleic acid sequences is a method for assaying “a material that itself has no specific and/or substantial utility.”

According to Anticipat Research, this type of rejection (Section 101 – Lack of Utility) has a very high reversal rate. Over the past two years, the PTAB has reversed 14 of 19 of these rejections, a relatively incredible 74% reversal rate. This may stem from the rejection being applied less frequently with other rejections and Examiners hence not having much experience with properly applying it.

With a reversal rate so low for lack of utility rejections, when struggling with an Examiner with such a type of rejection, it may be helpful to go straight to the Board on appeal to overcome such a rejection.

Board panel citing Berkheimer to reverse judicial exception rejection to diagnostics claims: no evidence

Since the two weeks since we predicted that the PTAB would start to dramatically change its outcomes of rejections under Section 101, we have seen no such change. Since then, recap emails have mostly shown affirmances (only 7 reversals of 86 total Section 101 decisions = 8% reversal rate). But a decision in yesterday’s recap email shows precisely the kind of rejection analysis that is expected to become more mainstream at the PTAB.

Ex Parte Galloway et al (PTAB May 22, 2018) reversed the judicial exception rejection under Section 101 because of a lack of evidence. The panel, consisting of Donald E. Adams, Demetra J. Mills, and Ulrike W. Jenks, found that the Examiner had not provided evidence to support a prima facie case of patent ineligible subject matter.

The panel cited to Berkheimer in support of an apparent defective step 2 analysis: “The Examiner has not established with appropriate factual evidence that the claimed method uses conventional cell counting methods.”

As a stylistic aside, Section 101 rejections are typically presented in decisions toward the very top of the document. It is unclear how or why (it may stem from examiners or practitioners ordering the statutory rejections), but this practice has gone on in the Board’s decisions for several years. However, a recent trend is for the Board to analyze Section 101 after prior art rejections. Now it makes sense why because a lack of a good prior art rejection can make for a good support that step 2 of a Section 101 rejection is improper.

And that is precisely what happened here. The panel proceeded to support its assertion (that step 2 of the Alice/Mayo framework was defective) by referring to its obviousness reversal. In other words, the Board’s finding of non-obvious claims supported that the claim features were not simply conventional or known in the art.

Another interesting point to note about this case is that it reinforces the much higher reversal rates of Section 101 judicial exceptions. The Board’s practice, as in this case, appears to be helping the patent-eligibility of diagnostics inventions.

As the PTAB becomes more confident in using Berkheimer in their decisions, expect more of the same analysis as Ex Parte Galloway. The appeal backlog has far too many cases where the Examiners did not have the guidance of Berkheimer to establish the proper evidence for Step 2. Thus, the necessary analysis from the Board need only be short and crisp.

Expect the Berkheimer-driven patent-eligibility pendulum to swing at the PTAB

The past few months have seen huge developments in patent-eligibility at the USPTO. In three and a half years after Alice, the most effective way to argue against patent-eligibility for software applications was to focus on Step 1–that the claims are not directed to an abstract idea. But based on these recent developments, Step 2–that additional elements of the claims transform the judicial exception into something more–looks to be the more powerful way. The only problem is that the PTAB has not yet caught on. It will.

These huge developments have taken place in the form of Federal Circuit decisions deciding patent-eligibility favorably to the patentee, especially Berkheimer v. HP Inc., 881 F.3d 1360, 1369 (Fed. Cir. 2018). Such a clear articulation of the need for factual findings for Step 2 should usher in big change in how the Alice/Mayo framework is applied.

Then on top of the decisions came the revised USPTO Berkheimer memo last month. These guidelines emphasized that to establish under Step 2 that an additional element (or combination of elements) is well-understood, routine or conventional, the examiner must find and expressly support a rejection in writing with one of the four:

1. A citation to an express statement in the specification or to a statement made by an applicant during prosecution that demonstrates the well-understood, routine, conventional nature of the additional element(s).

2. A citation to one or more of the court decisions discussed in MPEP § 2106.05(d)(II) as noting the well-understood, routine, conventional nature of the additional element(s).

3. A citation to a publication that demonstrates the well-understood, routine, conventional nature of the additional element(s).

4. A statement that the examiner is taking official notice of the well-understood, routine, conventional nature of the additional element(s).

It should come as no small surprise to any practitioner that Examiners have not been including the above support in their Step 2 analyses for these additional elements of claims. This is no slight to the examining corps; it simply was never a USPTO requirement. So if the PTAB were faithful to the principles set forth in the guidelines, one would expect a dramatic turning of the tide.

While the PTAB is not bound to the USPTO examiner memos, it shouldn’t stray too far from them. Plus, it must comply with the Federal Circuit decisions, which are consistent with the guidelines. So one wouldn’t expect the PTAB to continue its practice of overwhelmingly affirming on Section 101. However, so far the PTAB has not significantly deviated from its previous course of mostly affirming judicial exception rejections.

Since April 19, 2018–the day that the Berkheimer memo was published–there have been 120 decisions that have decided judicial exceptions. Of these, only 13 have reversed, meaning a reversal rate of 11%. This 11% reversal rate is below the recently reported reverse rate for abstract ideas of 14%. It would appear that panels have not yet had the time to incorporate this new Step 2 framework into their decision-making. Or alternatively, they are preoccupied with the arguments raised by the appellant. Expect a greater number of request for rehearings on these.

Sooner or later, these PTAB judges should realize that many Section 101 rejections on appeal do not have the proper support for Step 2. This is not to say that these Examiners, on remand, could reformulate a proper rejection given another opportunity. While theoretically the judges could affirm the 101 rejections with a designation of new, the Board may not be well-equipped for to do so as this new requirement requires factual basis supporting Step 2. That is, the PTAB is a body that decides the propriety of pending rejections, not a body for searching and making such support findings. So expect a greater number of reversals to let the Examiners follow Berkheimer.

Business method abstract idea rejections continue to get reversed at the Board, sparking form over substance debate?

In today’s Anticipat Recap email (March 22, 2018), three decisions at least partially reversed abstract idea rejections for business method claims. One of the decisions, Ex Parte Gershfang et al (PTAB Mar. 22, 2018), even included a dissenting opinion. Here, we report on an interesting decision pulled from another decision, Ex parte Beach (PTAB Mar. 22, 2018).

Before getting into Beach, some context behind the debate about form over substance of patent-eligibility. The Supreme Court in Alice v. CLS Bank struck down the method claims as abstract ideas, but also, importantly, struck down the computer system and computer readable medium claims for the same reason. The Supreme Court did not want clever practitioners to be able to get a claim based on the draftsman art. So just because a claim included computer recitations did not, by itself, make the claims more than an abstract idea. For this reason, it is very rare for the Board to split the claim set baby. For the most part, either all the claims are upheld as abstract or all the claims are reversed.

Ex parte Beach comes out differently than Alice because here the Board upheld the abstract idea rejection on the method claims but at the same time overturned the abstract idea rejection on the system claims. The representative method claim is here:

58. A method for processing coins, the method comprising:

receiving a plurality of randomly oriented coins in a coin counting machine;
dispensing a financial instrument from the coin counting machine in exchange for at least a portion of the coins;
receiving the financial instrument from a customer;
reading a code off the financial instrument with a machine;
sending at least a portion of the code to a database in a query for information to determine if the financial instrument is valid;
receiving the information from the database; and
redeeming the financial instrument for at least one of cash and merchandize [sic merchandise] when the information received from the database indicates the financial instrument is valid.

The Board found claim 58 to be a patent-ineligible abstract idea.

By contrast, the Board overturned the system claims as not an abstract idea. Here is a claim that was reversed:

91. A system for processing coins, the system comprising:

a coin counting machine, the coin counting machine having:

a coin input region configured to receive a plurality of randomly oriented coins from a user;
a coin discriminator configured to count the coins to determine a total; and
a dispenser configured to dispense a financial instrument to the user in exchange for at least a portion of the coins, wherein the financial instrument includes a code;

a database configured to receive information from the coin counting machine related to the financial instrument; and
a recognition subsystem configured to read the code off the financial instrument and send a related query to the database for information related to the financial instrument.

In its reasoning, the Board distinguished these system claims from the ineligible method claims because the structural elements of the system claims “simply do not recite an abstract idea.” According to the Board, these structural elements include a coin counting machine having “a coin input region configured to receive a plurality of randomly oriented coins from a user,” “a coin discriminator configured to count the coins,” “a printer configured to print a machine-readable code and a value on a voucher,” and “a dispenser configured to dispense the voucher.”

So here the Board holds that for a coin machine system claim, the structure of a coin input region, the coin discriminator to count the coins, the printer to print a voucher and the dispenser to dispense the voucher makes a claim patent-eligible.

Beach shows that the Board is willing to affirm some claims that were rejected as abstract ideas while reversing others. This is important because it shows that after Alice, the Board appears to be able to analyze the claims in a very much nuanced manner. Indeed, the extent that structural claim recitations can render a claim not abstract is far from over. But when it comes to a coin counting machine, these structural limitations were sufficient for the Board.

The Board will likely continue to render affirmed-in-part determinations on patent-ineligible judicial exception rejections only in rare cases. But we will report of any trends that develop.

PTAB finds encryption software claims not directed to abstract idea–Practice tips

How does the Board incorporate very timely case law in their decisions? How can you use this to your advantage? We consider these questions as we discuss a recent Board decision that reversed an abstract idea rejection while incorporating a very recent Federal Circuit into its analysis.

In a recent decision, Ex parte Jang (March 1, 2018) (available at https://anticipat.com/research?id=104015), the Board reversed an Examiner’s Section 101 rejection. In so doing, the Board panel found that the claimed software invention was not directed to an abstract idea under step 1 of the Alice/Mayo framework. Interestingly, the Board analogized to a Federal Circuit decision that had been decided not six weeks prior.

This recent Federal Circuit decision that the Board cited, Finjan, Inc. v. Blue Coat Sys., Inc., 879 F.3d 1299, 1303 (Fed. Cir. 2018), was not merely a cursory citation or a footnote. Rather, the Board specifically analogized to the patent in Finjan in holding that the claims at issue were similarly not directed to an abstract idea. How did the Board incorporate this legal authority into its own decision so quickly?

One thing that can be ruled out is any help from the appellant. Often times, the Board does indeed borrow persuasive arguments from the applicant (including persuasive legal authority). At times, you can see arguments literally lifted from appellants’ appeal briefs or reply briefs, as discussed in this blog post. Not so here. The reply brief was filed in September of 2016, long before Finjan was decided. Plus, the appellant choose not to pursue an oral hearing, which would have given the appellant a chance to address more recent case law. See 37 CFR 41.47 (“Upon a showing of good cause, appellant and/or the primary examiner may rely on a new argument based upon a recent relevant decision of either the Board or a Federal Court.”).

Instead it appears that PTAB judges keep very current on Federal Circuit case law and use it in their own decision drafting. This is somewhat different from examiners, who, in the face of a recent Federal Circuit decision, may have problems addressing the applicability of recent Federal Circuit decision. The Examiners may instead prefer the USPTO to interpret any new Federal Circuit decisions in an official USPTO guideline.

 

Now that we know that the Board analogizes to recent legal authority, how is this useful? How can you use the fact that the Board regularly cites to recent and relevant case law in reversing Examiner rejections? It turns out that this is most useful in fast-evolving areas of law, such as Section 101. For example, you can cite to the same legal authority that the Board uses to overturn your own Examiner rejections.

As we pointed out in a recent blog post, you can input any application number into the Anticipat Practitioner Analytics input field, and it will show the relevant legal authority for this application. Let’s say that you have an application in the software arts that is meeting stiff Section 101 headwinds. Perhaps, your application is even in the technical space of encryption. Let’s even say art unit 3621 (the same art unit that Ex parte Jang was decided).

After inputting art unit 3621 into the Analytics input field, you will see that the art unit field has legal authority for step 1 of the Alice/Mayo framework. There are three decisions where the Board relied on step 1 of the abstract idea test to reverse the rejection.

artunit3621

Within step 1, you see legal authority that the Board uses to support its reversal by the icon to the immediate right of the tag. This icon, which is clickable, provides all the decisions (and legal authority) for this art unit. It shows this popup for this art unit:

artunitlegal.png

In addition to Finjan, this Board decision used two other Federal Circuit cases in reversing the rejection: Enfish and Thales Visionix.

Chances are that if you have an application in this art unit and are facing a similar stubborn rejection, accessing legal authority that the Board found worthy of overturning an examiner can help guide strategy for what is persuasive. At best, the relevant legal authority is also found to be analogous and persuasive in your case. At worst, you are not completely off-base in citing to such a decision since independent judges in this art unit analogized to such legal authority.

The next column, tech center, has even more decisions and legal authority. Finally, the right-most column displays all the tags for this specific tag. This will provide all the legal authority Board decisions.

This practice tip is well-suited for responding to Office Actions as responses under 1.111 because you can quickly use recent legal authority. In theory, this same legal research could guide an applicant preparing for an appeal. This is especially relevant since if judges are using specific legal authority to overturn specific rejections, chances are they will be receptive to such legal authority in your brief. However, due to the substantial delay in reaching a final decision, and due to the volatility of Section 101, some of the cases may be outdated by the time the panel decides your case.

See what Practitioner Analytics can do for your practice. Feel free to give Analytics a 14-day free trial.

Let us know if you have any questions or if you would like a demo.

Anticipat Team