Archive for the ‘Uncategorized’ Category

EPO Provides Patentability Guidance for AI-based Applications

Monday, October 22nd, 2018

The European Patent Office (EPO) has provided further guidance for examination in relation to the patentability of inventions involving mathematical methods and computer programs. This updated guidance is of particular relevance to inventions relating to the fast-growing field of Artificial Intelligence (AI).  This guidance was published as part of the EPO’S annual update of the “Guidelines for Examination”.

The new EPO Guidance first defines AI and ML as being “computational models and algorithms for classification, clustering, regression, and dimensionality reduction, [and which may include] neural networks, genetic algorithms, support vector machines, k-means, kernel regression, and discriminant analysis.”  Additionally, the Guidance states that such computation models and algorithms relating to AI and ML are “per se of an abstract mathematical nature,” indicating that the EPO will likely treat such algorithms as unpatentable by default.
Generally, under examination by the EPO, applications involving mathematical methods are excluded from patentability unless they are determined to have technical character.  In assessing whether a mathematical method possesses such technical character, a determination is made whether the invention produces a technical effect that serves a technical purpose.  A generic purpose such as “controlling a technical system” is not sufficient to confer technical character to the mathematical method.  However, the Guidance specifically notes that “artificial intelligence and machine learning find applications in various fields of technology,” and highlights examples of a “neural network in a heart-monitoring apparatus” and “classification of digital images, videos, audio or speech signals based on low-level features” as both possessing technical character.  In contrast, the EPO identified the classification of text documents solely based on their textual content and classification of abstract data records without any indication of a particular technical use as not having technical purpose.  Furthermore, the EPO treats expressions such as “support vector machine”, reasoning engine”, or “neural network” as merely referring to abstract models that are “devoid of technical character.”
Going forward, it appears that AI-ML-related applications filed in the EPO should specifically highlight how a specific field of technology is improved by the AI-ML-related mathematical methods in order to best demonstrate technical character.
My thanks to Greg Rabin from Schwegman, Lundberg & Woessner, P.A., for this post.

PTAB Reverses 101 Rejection for Ranking Based on Web Page Dwell Time

Friday, October 12th, 2018

On an appeal from a 101 rejection by an examiner, the PTAB has reversed a 101 rejection to the following claim, from eBay’s U.S. Application No. 12/814,020:

18. A method comprising:

identifying a plurality of listings stored in a listing database as search results;

determining, using a processor, a respective dwell time associated with each of the plurality of listings, the dwell time based on an elapsed amount of time one or more users view a page describing the listing, and the dwell time associated with a likelihood of a transaction occurring with respect to the listing; and

ranking the listings composing the identified plurality of listings based at least in part on the respective dwell time associated with each of the plurality of listings.

The PTAB reasoned:

“We determine claim 1 is directed not only to the “business idea” of organizing (ranking) search results, but claim 1 also is directed to the use of dwell time, which is an Internet-centric challenge. See, e.g., Spec. 1–4, 12-16. Claim 1 is analogous to the claim at issue in DDR Holdings in that, in both cases, a key aspect of the claim focuses on “a challenge particular to the Internet.” DDR Holdings, 773 F.3d at 1257.”

I agree with the PTAB that this claim involves a idea related to the operation of a computer system, not an idea where novelty is based on a new piece of information or a new type of information that is abstract in nature, and therefore should fall inside 101.

The brief for this appeal was written by Schwegman, Lundberg & Woessner, PA, and in particular, by Jeff Ranck.



eBay is attempting to patent a list of listings ranked by “dwell time” — the “elapsed amount of time one or more users view a page describing the listing.”

Recent PTAB Reversals of Section 101 Rejections

Monday, October 8th, 2018

Here are some recent PTAB Reversals of Section 101 rejections:

Ex parte Oudenallen—Claims for a garment for preventing hypothermia:

The PTAB Reversed an Examiner’s 101 Rejection of Claims in a Facebook Patent Application, Citing Berkheimer:

An IBM patent application saved by Berkheimer:

WMS Gaming/Bally case —a Win for WMS at the PTAB under 101:

The PTAB Reversed a 101 Rejection of Claims in a Halliburton Patent Application:

The PTAB Reversed an Examiner’s 101 Rejection of Claims for a Method for Measuring Central Aortic Pressure:

The PTAB Reversed an Examiner’s Rejection of Claims for Adhesive Labels,; Yes, that’s right, Adhesive Labels:




Recent PTAB 101 Decisions: The Good, The Bad, and The Ugly

Tuesday, September 25th, 2018

My thanks to Janal Kalis of Schwegman, Lundberg & Woessner, P.A., for this post:


The PTAB reversed an examiner’s 101 rejection of claims in an SAP patent application.  The PTAB also reversed on 103.  However, there is a dissenting opinion:

The PTAB Reversed an Examiner’s 101 Rejection of Claims in a CalTech Patent Application for a Method for Predicting 3D Structures, such as Proteins:


The PTAB Affirmed an Examiner’s 101 Rejection of Claims for an Information Display Device But Reversed the Examiner’s 102 Rejection–this is what Dir. Iancu spoke of yesterday:

The PTAB Affirmed an Examiner’s 101 Rejection of Claims in an IBM Patent Application But Reversed the Examiner’s 103 Rejection–How many more of these must we see:


The PTAB Affirmed an Examiner’s 101 Rejection of Claims for an Electrostatic Clamp Monitoring System:

Another Outrageous Decision from the PTAB Affirming an Examiner’s 101 Rejection of Claims in a Becton Dickenson Patent Application for Measuring the Amount of Blood in a Culture Vessel:

PTAB Reverses 101 Rejection on Authentication Invention for MasterCard

Friday, September 14th, 2018

In a recent PTAB appeal, Ex parte MICHAEL C. WARD, PATRIK SMETS, and PAUL VANNESTE (Appeal 2017-000625/Application 12/833,059) the Board reversed a 101 rejection against a patent application owned by MasterCard International, Inc.  Claim 1 is recited below.

Instead of getting into the two-step Alice test, I thought it might be instructive to look at this invention from the simplistic point of view of whether the invention was functional in nature, like a music player, or only like music, the former being technology that should be patentable, and the latter not.

One way to analogize the claim in question here is to say that the data processed and used by the system is analogous to “music”  – in particular the first and second cryptograms, and the extra data.   This data is processed using a method that can be analogized to the operation of an software-implemented music player – music data is the input, and sound is the output.   The question is then whether there is an innovative aspect to this method (i.e., the music player), or is only the data (i.e., the music) innovative.

In the instant case, the method in essence requires the following:

  • obtaining, at a terminal from a payment device reader, a first and a second crytogram
  • transmitting, from the terminal, the first and second cryptograms in a first message, and the extra data, through a payment network
  • obtaining, at the terminal, a second message corresponding to authentication of the payment device
  • wherein the authentication is issued upon a first cryptographic calculation upon determining that the first message and extra data have been obtained by the issuer
  • wherein the cryptographic calculation comprises:
    • running a first message authentication code calculation using said extra data
    • running a second message authentication code calculation using said extra data; and determining said authentication of said payment device by comparing a truncated portion of an output of each of said first and second message authentication code calculations to said first cryptogram and said second cryptogram

By analogy to the music and music player, the method as claimed specifies that the data in question is input into a system, processed in a non-obvious way, and to generate an authentication output for the payment device.  So, in this instance, while there is music and a music player by analogy, and even assuming the cryptograms and second data are unpatentable data, like unpatentable music, there is still novelty in the methodology used to process this data, to achieve the authentication of a physical payment device.  Therefore, this is not an attempt to predicate patentability on abstract data alone, but provides a mechanism to authenticate a physical device using a novel processing methodology.

Claim 1 of U.S. Application No. 12/833,059:

1. A method comprising the steps of:

obtaining, by a terminal component from a payment device reader component, at least a first cryptogram and a second cryptogram;

transmitting, from said terminal component to an issuer of a payment device presented to said payment device reader component, through a payment network, said first cryptogram, said second cryptogram, and extra data, wherein at least said first cryptogram and said second cryptogram are transmitted in a first message;

and obtaining, by said terminal, a second message from said issuer, said second message corresponding to authentication, by said issuer, of said payment device presented to said payment device reader component, said authentication being issued upon a first cryptographic calculation, wherein said first cryptographic calculation is selected from among a plurality of cryptographic calculations upon determining that said first message and said extra data have been obtained by said issuer, wherein said first cryptographic calculation comprises:

running a first message authentication code calculation using said extra data;

running a second message authentication code calculation using said extra data; and determining said authentication of said payment device by comparing a truncated portion of an output of each of said first and second message authentication code calculations to said first cryptogram and said second cryptogram.

Is Machine Learning Technology Going to be Collateral Damage under the Abstract Idea Exception?

Monday, August 6th, 2018

My colleague Janal Kalis identified 21 PTAB decisions directed to machine learning patent applications published since 4/1/2018.  According to Janal, seventeen of the patent applications considered had been rejected by examiners with 101. The PTAB affirmed examiners’ 101 rejections in 16 cases and reversed in only 1 Case.

Here is a list of the cases:  PTAB Decisions in Machine Learning Cases (1844172)

It might be a good time to take a step back and see whether these rejections are consistent with sensible policy, even if justifiable under existing 101 jurisprudence, so at least we can start to assess the potential long term damage these decisions will do to innovators seeking protection in the US for this important technology.

Is the Abstract Idea Exclusion A Big Mistake for Software, or is There Something Important To It?

Sunday, August 5th, 2018

As recently reported by Dennis Crouch (, under Dir. Iancu, eligibility rejections have dropped to about 6.6%, from approximately 8% before that. Software patent practicioners are pleased to see the needle moving in the “right” direction, but still there are far too many worthy inventions being denied by the USPTO, the PTAB, and ultimately the Federal Circuit. In particular, inventions based solely on highly innovative and 100% technical, electrical engineering subject matter, continue to be treated as if they were claiming methods for practicing voodoo. Let me add that many if not even most of these rejections are perfectly justifiable based on the most recent Supreme Court 101 jurisprudence, and the Federal Circuit jurisprudence attempting to implement those flawed decisions. And therein lies the crux of problem currently faced by the patent system in dealing with the eligibility of many software-related inventions: both the public policy and the logic behind this jurisprudence is a disaster.

I was hoping that eventually someone would come forward with a clear way to analyze inventions to determine if they are “abstract” or “not abstract.” But so far, no one has done it. I have tried to puzzle it through to a logical conclusion, but have not gotten there yet, or perhaps not even close. But I do have some observations that help me understand the nature of what is “abstract” or “not abstract.”

First, l believe that a large majority of inventions submitted for patenting at the USPTO start out their existence, and end their existence, purely in the human imagination. I say a large majority and not all, because in some cases no doubt an invention is physically created simultaneously with its emergence in the inventor’s mind, such as a mechanical invention made possible due to the placement of an additional part in a combination, where the part is physically positioned simultaneously with the inventor’s decision to try it there. But certainly very often, inventions are only imagined by an inventor and the only existence they obtain is perhaps being described in a patent application or inventor’s notebook.

Are these inventions that exist only in the imagination (even when documented they still only have meaning when imagined by the reader) “abstract” so as to not be patent eligible? Of course we know an invention, even if never “actually reduced to practice” is considered “constructively reduced to practice” if it is described in a patent application. So, this cannot be what is meant by “abstract”, in and of itself. So, it is possible for an invention to only exist in an “imagined” form, not be “abstract”, and still be patent eligible. But it is also possible that an invention that is only “imagined” can also be “abstract” under Section 101.

In a different realm of consideration, I often like to entertain the idea that a novel musical score is a good example of what we could consider “abstract” under Section 101. More particularly, we know most any good patent attorney could draft a patent application that described a “novel” computer configured to generate sound energy in accordance with the musical score, using a computer program. By traditional eligibility logic applied to computer-implemented inventions, this would constitute a uniquely configured computer, and certainly the computer as a whole in combination with the unique program, are as “technical” as you can get. But, most practicioners would never have even attempted to patent a new musical score in the form of a programmed computer, absent some other engineering innovation that was not limited to the novel score. Why? Because,this hypothetical “invention” is not technology, but rather the work of a composer. Note, strictly speaking, this “invention” is no more abstract than any other software invention that effectuates novel operation of a computer — it is just a different type of functionality — one that is not scientific in nature by historical conventions.

On a related note, consider the printed matter doctrine — e.g., see In re Gulack (Fed Cir 1983), which holds that printed matter is not patentable unless it has a “functional relationship to the substrate.” In essence, in the example above, the novel subject matter, the musical score, does not in and of itself have an “functional relationship” to its “substrate – i.e., the computer. Accordingly, the computer cannot be said to be “part of” the invention.

So, for today, I will leave you with this idea: could it be that “abstract” under 101, as applied to software, mean at least in part, that the novel aspect of the invention at hand, has no functional relationship to the computer hardware itself, such that the hardware is not fairly part of what is being claimed? This is not to say that there is not technical subject matter being included in such a claim, but if that technical subject matter is not truly functionally integrated with the novel subject matter, perhaps the novel subject matter is indeed merely “abstract” under the printed matter doctrine.  Could it be that some computer-related inventions are nothing more than novel arrangements of information more akin to a work of authorship than a technical invention, and that the computer only, essentially, “plays” that information the way a computer might play a new song using a music app?   Or is the computer just functioning essentially as a calculator, performing mathematical operations that are not functionally related to the calculator itself.

Please understand these thoughts are by no means claimed by me to be worthy of application in practice, so I am not arguing for their adoption.  I am merely trying to better understand if there is any solid, logical basis, for excluding some types of software-implemented “inventions” from eligibility.

Ex Parte Jadran Bandic, et al. — The bar is raised for proving the presence of an abstract idea

Wednesday, May 9th, 2018

The USPTO recently provided a Memorandum and Examiner training related to a Federal Circuit decision, Berkheimer v. HP Inc., 881 F.3d 1360 (Fed. Cir. 2018). The Memo and training give hope that the patent office is taking seriously the need for consistency in applying rejections under 35 U.S.C. §101. While the Memo and training apply to Examiners, it remains an open question how the Patent Trial and Appeal Board (PTAB) will consider the contents of the Memo.

In Ex Parte Jadran Bandic, et al., (Decision) decided by the PTAB on April 30, 2018, this question appears to be partially answered. The case hinges on a §101 analysis of a single independent claim with five steps. The PTAB first reverses the Examiner’s rejection under Alice Step One as failing to consider the claim as a whole, stating that the “The Examiner’s asserted abstract idea is based on “the steps within the [claimed] method of comparing and determining” and ignores the claim as a whole” (Decision, pages 4-5). Despite the sufficiency of the Step One reversal for disposing of the Appeal, the PTAB also considered the Alice Step Two analysis.

In the Alice Step Two section of the Decision, the PTAB found that “The Examiner’s analysis of the second step of the Alice analysis is conclusory and unsupported” (Decision, page 6). The PTAB continued by making it clear that unsupported statements from an Examiner are insufficient for complying with Step Two (See, e.g., Decision, pages 5-6). This sentiment is certainly suggested and supported by the Berkheimer case, but more importantly, the analysis by the PTAB here upholds the requirements presented in the Memo that are placed on Examiners in making a Step Two rejection. Namely, that absent a citation to an Applicant’s admission, a court decision, or a publication “demonstrating the well-understood, routine, conventional nature of the additional element(s),” a rejection under Step Two is inappropriate (Memo, pages 3-4).

Thus, the Decision validates the Memo’s interpretation of Berkeheimer requiring Examiners to “expressly support[] a rejection in writing” with a citation to an Applicant’s statement, court decision, or publication, and further that the citation must demonstrate not only that the identified additional elements were known, but well-known, routine, and conventional.

My thanks to Jeff Cobia of Schwegman, Lundberg & Woessner, P.A., for this posting.

USPTO Posts slides for new examiner training on Section 101

Monday, May 7th, 2018

On May 7, 2018, the USPTO posted the slides for the Examiner Training in relation to the Berkheimer decision, Berkheimer v. HP Inc., 881 F.3d 1360 (Fed. Cir. 2018), titled Subject Matter Eligibility: Well-Understood, Routine, and Conventional Activity.  The slides start off by making it clear the Berkheimer decision does not change the basic subject matter eligibility framework, but does clarify how to determine if an element(s) represent well-understood, routine, or conventional activity in step 2B of the subject matter eligibility analysis.  The slides distinctly point out that obviousness or lack of novelty does not establish that the additional elements are well-understood, routine, or conventional activities.


As seen in the Berkheimer Memorandum, the examiners are now instructed to expressly support a rejection with one of the following four options.  Option 1: A statement by the applicant in either the specification or during prosecution.  Option 2: a citation to a court decision discussed in MPEP § 2106.05(d)(II).  Option 3: a publication, with the slides including a reminder that merely finding the additional element in a single patent or application would not be sufficient to establish the element as well-understood, routine, or conventional, unless that patent or application demonstrates that the additional  element is widely prevalent or in common use in the relevant field.  Option 4: the examiner may take official notice.  The slides strongly caution using this option and state it should be used only when the examiner is certain the additional element is well-understood, routine, and conventional.


The examiners were reminded to consider the additional elements individually and in combination, and thus to support a rejection of a claim where the examiner takes the position that two elements are routine, then the combination of those two elements must also be shown to represent well-understood, routine, and conventional activity.  When the examiner’s position is challenged by the applicant, the examiner should reevaluate whether it is readily apparent that the additional elements are in actuality well-understood, routine, conventional activities to those who work in the relevant field.  If the applicant is responding to when the examiner has taken official notice, then the examiner should reevaluate  and provide support from options 1-3.

My thanks to David Dyer at Schwegman, Lundberg & Woessner, P.A., for this post.

USPTO to Issue new 101 Guidance Today!

Thursday, April 19th, 2018

Very much looking forward to the new guidance on 101 from the USPTO.  I am very encouraged that the USPTO is going to take a leadership role in leading our country out of the 101 “Alice in Wonderland” era.  Software 101 jurisprudence is probably the most absurdly vague, inscrutable and unpredictable body of law in the entirety of the US legal system.  Worse yet, the weakening of the patent system has done absolutely nothing to help start ups.  On the contrary, it has most likely accelerated the decline of tech start-ups.  Ultimately, that can’t be good news for the U.S.

So, let’s hope the USPTO can help the U.S. put the Section 101 chaos of the last few years behind us once and for all.