Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
DETAILED ACTION
Applicant filed an amendment on 9/14/26. claims 1-20 were pending and stand rejected. Applicant amends claims 1-2, 7, 9-10, 15-17. Claims 6, 8 and 14 are cancelled.
After careful consideration of applicant arguments and amendments, the examiner finds them be moot and/or non-persuasive, this action is a Final Rejection.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1,9 and 16 are is/are rejected under 35 U.S.C. 103 as being unpatentable over US Patent Publication to Livesay 20220374847 in view of US Patent to Murdock 10169717
As per claims 1, 9 and 16 Livesay discloses; A system comprising: a tangle network comprising a plurality of computing nodes connected to a network,
wherein the tangle network stores a tangle comprising a directed acyclic graph (DAG) of a plurality of nodes; Livesay(0059-60 tangle Acyclic graphic)
and at least one processor communicatively coupled to the tangle network and configured to: monitor a plurality of data interactions; Livesay(0145, monitor the ledger network… generally data interaction)
identify a plurality of potential exceptions relating to the plurality of data interactions,
Livesay(0170 monitor for human error)
wherein each potential exception relating to a particular data interaction of the plurality of data interactions indicates a possible error associated with reporting that particular data interaction to an entity; Livesay(0018)
analyze each potential exception based on a respective pre-configured logic to determine a result relating to that potential exception, wherein the result indicates whether that potential exception is a valid exception; Livesay(0097 recognize patterns in the identified data)
generate a plurality of data objects, wherein each data object comprises one or more of the potential exceptions, results determined for the potential exceptions, pre-configured logic used to analyze the potential exceptions, (Livesay, including recognize deviations 0098, deviation from patterns to identify a likelihood of fraud)
or data used in analyzing the potential exceptions based on the respective pre-configured logic; pass the plurality of data objects as nodes of the tangle associated with the tangle network, wherein each data object is passed to the tangle as a separate node; deploy a federated machine learning model associated with each node passed to the tangle to determine whether the results determined for the potential exceptions are correct, (objection) Livesay(0018, use tangle network and federated learning model to analyze the data)
wherein each federated machine learning model associated with a particular node of the tangle uses data from the particular node and previous two nodes of the tangle to determine whether the results determined for the potential exceptions are correct,
Livesay(0029 multi layer approach)
wherein determining whether a result associated with a potential exception is correct comprises: when the result indicates that the potential exception is a valid exception, determining whether the result correctly identifies the potential exception as a valid exception; and when the result indicates that the potential exception is an invalid exception, determining whether the result correctly identifies the potential exception as an invalid exception;
wherein deploying the federated machine learning model associated with each node comprises:
Livesay(0129)
deploy a first federated machine learning model configured to receive first input data from a first node comprising a first data object and at least two previous nodes of the tangle, the first federated machine learning model determining a first confidence indicator based on the first input data, wherein the first confidence indicator indicates a degree of confidence that the results determined for the potential exceptions are correct;
Livesay(0170, looks for errors to correct or not and corrects them, in regards to federated models, the definition is that the data is kept separate in different databases and analyzed separately, it appears in fig. 4 that this is how Livesay works)
and generate a report comprising a plurality of the potential exceptions determined as valid. Livesay (0094 output evaluations of the data)
determine one or more remedial measures to correct valid exceptions identified in the report; and execute the one or more remedial measures to correct the valid exceptions, wherein executing a remedial measure comprises re-sending a corrected reporting of a data interaction associated with a valid exception.
Livesay(0170, correct and clear, typographical elements for example, “remedial” could be a repair, one or more could be just one)
Here Livesay appears to be able to use a tangle network to take remedial actions and find errors for collection. However, in regards to having multiple as in three levels of hierarchical confidence levels Livesay does not explicitly disclose what Murdock teaches;
deploy a second federated machine learning model configured to receive second input from a second node comprising a second data object, the first node, and a node previous to the first node of the tangle, the second federated machine learning model determining a second confidence indicator based on the second input data, wherein the second confidence indicator is higher than the first confidence indicator and indicates with a higher degree of confidence that the results determined for the potential exceptions are correct; and in response to determining that the second confidence indicator equals or exceeds a threshold: determine that the results determined for the potential exceptions are correct; and
deploy a third federated machine learning model configured to receive third input data comprising a third data object from a third node comprising a list of the potential exceptions, the second node, and the first node, wherein the third input data received from the third node comprises a third data object comprising a list of the potential exceptions, and wherein the third federated machine learning model determines, for each potential exception in the list of potential exceptions in the third data object, whether the respective result determined for the potential exception is approved or rejected;
Here, in regards to three levels of answers, Murdock teaches using three sources to obtain a better answer and so is provided for that element using confidence analysis to compare confidence and obtain the best answer. (col. 11-col. 15)
It would therefore have been obvious to one of ordinary skill in the art before the effective filing date of the invention to combine the federated teachings and tangle network of Livesay and the three node comparison of results of Murdock for the motivation of “a collaborative model” to provide answers. (Col. 1 lines 5-20)
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 2-5,7, 10-13,15 and 17-20 are is/are rejected under 35 U.S.C. 103 as being unpatentable over US Patent Publication to Livesay 20220374847 in view of US Patent 10169717 to Murdock and further in view of US Publication to Mahr 20190273617
As per claims 2, 10 and 17 Livesay and Murdock do not explicitly disclose what Mahr Teaches;
The system of Claim 1, wherein the at least one processor is configured to generate the plurality of data objects by: generating a first data object comprising a list of the potential exceptions and an indication of whether each exception in the list was determined to be valid; generating a second data object comprising the pre-configured logic and data used to determine whether each potential exception is valid; and generating a third data object comprising a list of potential exceptions relating to which a decision is to be made regarding whether the results determined for the potential exceptions are correct, wherein the decision relating to each potential exception indicates whether a respective result determined for the potential exception is correct. Mahr(0162, exceptions, revocation lists 0216, 0270 list of bindings) It would therefore have been obvious to one of ordinary skill before the effective filing date of the invention to combine the Tangle network and DAG disclosure of Livesay with the exception/revocation teachings of Mahr for the motivation of efficiently serving customers. (0025)
As per claims 3, 11 and 18 Livesay discloses;
The system of Claim 2, wherein the at least one processor is configured to pass the plurality of data objects to the tangle by: passing the first data object to the tangle as a first node of the tangle; after passing the first data object to the tangle, passing the second data object to the tangle as a second node of the tangle; and after passing the second data object to the tangle, passing the third data object to the tangle as a third node of the tangle.
Livesay (0059-60 this is how a tangle network works)
As per claims 4, 12 and 19 Livesay discloses; The system of Claim 3, wherein the at least one processor is further configured to: deploy a first federated machine learning model that is configured to receive first input data from the first node and at least two previous nodes of the tangle; deploy a second federated machine learning model that is configured to receive second input data from the second node, the first node and a node previous to the first node of the tangle; and deploy a third federated machine learning model that is configured to receive third input data from the third node, the second node and the first node.
Federated (fig. 16)
As per claims 5, 13 and 20 Livesay and Murdock doe not explicitly disclose what Mahr teaches;
The system of Claim 4, wherein the at least one processor is further configured to: iteratively run the first federated machine learning model, the second machine learning model and the third machine learning model one after the other to determine whether the results determined for the potential exceptions are correct. Mahr(0162, ie. Exceptions, good or bad)
It would therefore have been obvious to one of ordinary skill before the effective filing date of the invention to combine the Tangle network and DAG disclosure of Livesay with the exception/revocation teachings of Mahr for the motivation of efficiently serving customers. (0025)
As per claims 7 and 15 Livesay discloses; The system of Claim 6, wherein the at least one processor is further configured to: run the third federated machine learning model to determine a third confidence indicator based on the third input data from the third node, the second node and the first node.
Livesay(fig. 16, federated, nodes fig. 9)
Potentially allowable; Claims 7, 15 combined with 5, 13 which depend respectively on 6, 5, 4, 3,2, (respective chain for 7) if rolled up, consistent with the offering provided after non-final.
Response to Arguments
Applicant filed an amendment on 9/14/26. claims 1-20 were pending and stand rejected. Applicant amends claims 1-2, 7, 9-10, 15-17. Claims 6, 8 and 14 are cancelled.
After careful consideration of applicant arguments and amendments, the examiner finds them be moot and/or non-persuasive, this action is a Final Rejection.
Claims Objections- moot
Claim Rejections under 35 U.S.C. 101- moot in view of amendment and applicant arguments that this invention is a mental process of machine learning that cannot be conducted in the mind.
Claim Rejections under 35 U.S.C. 102- moot in view of amendment resulting in new grounds of rejection.
Claims 1, 8, 9 and 16 are rejected in view of US Patent Publication No. 2022/0374847 ("Livesay").
Applicant respectfully requests reconsideration and withdrawal of these rejections.
Applicant traverses this rejection and respectfully submits that Livesay fails to teach each and every element of the these claims. For example, Livesay fails to teach, suggest, or disclose "deploy a federated machine learning model associated with each node passed to the tangle to determine whether the results determined for the potential exceptions are correct..." as recited in amended claim 1.
The Examiner appears to conflate "federated distributed ledger" of Livesay with "federated machine learning". The error in the Examiner's rejection is the conflation of two entirely unrelated technologies that happen to share the word "federated." Claim 1 recites, in part, "deploy a federated machine learning model associated with each node passed to the
tangle to determine whether the results determined for the potential exceptions are correct..." The Examiner cites Livesay paragraph [0018], characterizing it as disclosing "use tangle network and federated learning model to analyze the data." See Office Action at page 6. This characterization is incorrect. Paragraph [0018] of Livesay discusses "title defects" and "Human Induced Defects (HIDs) which often come in the form of mistyped names, transcription errors from paper-based documents and other non-automated sources, and missing property descriptions or incorrect legal descriptions." Paragraph [0018] does not mention "federated learning," "federated machine Learning," or any machine learning model of any kind. The phrase attributed to Livesay by the Examiner, "use tangle network and federated learning model to analyze the data", does not appear in paragraph [0018] or anywhere else in Livesay.
The term "federated" in Livesay appears exclusively in the context of a "federated distributed ledger layer," which is a permissioned blockchain architecture. Specifically, Livesay states: "The multi-layered hybrid distributed ledger architecture also includes a federated distributed ledger layer which requires nodes to receive permission to append data to the federated distributed ledger layer." Livesay at paragraph [0029]. A federated distributed ledger is a type of permissioned blockchain where multiple authorized entities can read and write data to the blockchain. This is a fundamentally different technology from "federated machine learning," which is a machine learning architecture in which a model is trained across multiple decentralized devices or servers holding local data samples without exchanging those data samples. These are distinct concepts from different domains of computer science. The Examiner's mapping is based on a conflation of two unrelated technologies that share only the word "federated," and this error pervades the entire rejection.
Livesay further fails to teach, suggest, or disclose "a tangle network comprising a plurality of computing nodes connected to a network, wherein the tangle network stores a tangle comprising a directed acyclic graph (DAG) of a plurality of nodes" as recited in amended claim 1.
The Examiner cites Livesay paragraphs [0059]-[0060] for "tangle Acyclic graphic." See Office Action at page 5. While Livesay does state that "[d]istributed ledgers may also include a tangle, a block lattice, or other directed acyclic graph (DAG)" at paragraph [0059], this disclosure is a single sentence in an exhaustive enumeration of possible distributed ledger types. Livesay's entire operative system is a "multi-layered hybrid distributed ledger518027330.1
architecture" that utilizes blockechain layers, public, federated, and private, for managing property records. The tangle is never the operative data structure in Livesay but is mentioned only in passing as one possible alternative form of distributed ledger. A bare mention of a tangle as one of several possible ledger types, without any description of how such a tangle operates or is implemented in the context of the disclosed system, does not satisfy the exacting standard of anticipation under Section 102. Under that standard, the identical invention must be shown "in as complete detail as is contained in the ... claim."
Here in the context of the federated argument, a definition of federated machine learning is that the data is kept separately. This speeds up the process and keeps the data separate. Ie as in the words of the attorney, “trained across multiple decentralized devices or servers holding local samples”…. Thus fig. 4 of Livesay shows this.
Livesay further fails to teach, suggest, or disclose "pass the plurality of data objects as nodes of the tangle associated with the tangle network, wherein each data object is passed to the tangle as a separate node" as recited in amended claim 1. The Examiner provides no specific citation for this limitation. This is a critical limitation that appears to have been entirely overlooked. Livesay's operative distributed ledger records property transactions in blocks on a blockchain, not data objects posted as individual nodes on a tangle. Under MPEP § 2143.03, "[a]tl words in a claim must be considered in judging the patentability of that claim against the prior art." The Examiner's failure to address this limitation renders the rejection legally insufficient.
Livesay further fails to teach, suggest or disclose, "wherein each federated machine learning model associated with a particular node of the tangle uses data from the particular node and previous two nodes of the tangle to determine whether the results determined for the potential exceptions are correct" as recited in amended claim 1. The Examiner cites Livesay paragraph [0029] for "multi layer approach." See Office Action at page 6. Paragraph [0029] describes "a federated distributed ledger layer which requires nodes to receive permission to append data to the federated distributed ledger layer." This describes layers of a permissioned blockchain architecture. This has nothing to do with a machine learning model that consumes data from three sequential nodes of a tangle (the current node and the preceding two nodes). The claim recites a specific data-flow topology in which each federated ML model looks at exactly three tangle nodes; this precise architectural requirement has no analog anywhere in Livesay.
Here in regards to the tangle network, Livesay teaches this. However in the view of applicant argument re: three nodes, the examiner is applying Murdock in combination. It is noted that “exactly three tangle nodes” does not appear to be claimed in this manner. Thus three is claimed, not “exactly” which would be more limiting since more than three would not be applicable.
Livesay additionally fails to teach, suggest, or disclose the following amended claim elements from amended claim 1:
wherein deploying the federated machine learning model associated with each node comprises:
deploy a first federated machine learning model configured to receive first input data from a first node comprising a first data object and at least two previous nodes of the tangle, the first federated machine learning model determining a first confidence indicator based on the first input data, wherein the first confidence indicator indicates a degree of confidence that the results determined for the
potential exceptions are correct;
deploy a second federated machine learning model configured to receive second input from a second node comprising a second data object, the first node, and a node previous to the first node of the tangle, the second federated machine learning model determining a second confidence indicator based on the second input data, wherein the second confidence indicator is higher than the first confidence indicator and indicates with a higher degree of confidence that the results determined for the potential exceptions are
correct; and
in response to determining that the second confidence indicator equals or exceeds a threshold:
pote
PNG
media_image1.png
24
630
media_image1.png
Greyscale
deploy a third federated machine learning model configured to receive third input data comprising a third data object from a third node comprising a list of the potential exceptions, the second node, and the first node, wherein the third input data received from the third node comprises a third data object comprising a list of the potential exceptions, and wherein the third federated machine learning model determines, for each potential exception in the list of potential exceptions in the third data object, whether the respective result determined for the
potential exception is approved or rejected;
Livesay's system uses machine learning elements to "recognize deviations from patterns to identify a likelihood of fraud". See Livesay at paragraph [0098]. However, Livesay does not teach deploying separate, sequential federated ML models, each consuming data from a
particular tangle node plus the previous two nodes, where each model produces an escalating
confidence indicator and where the system only proceeds to the next stage when the confidence indicator meets a threshold. This is a fundamentally different architecture. Livesay uses ML in a general-purpose pattern recognition context for property fraud detection, whereas the claimed
invention implements a gated, iterative confidence-escalation pipeline tied to a tangle network's data-access properties. Maher (US Patent Publication No. 2019/0273617) cited to reject dependent claims does not employ machine learning at all. Maher uses cryptographic hashing, policy engines, and Byzantine consensus protocols for trust assertions. The concept of an escalating confidence indicator derived from federated ML models is wholly absent from Maher.
Here the applicant arguments are in part Moot in view of Murdock. It is noted that “separate sequential federated machine learning models” is likewise not claimed or it would also be limiting and thus potentially allowable in the context of the limits set by the claim.
Claims 9, 16 are similar to claim 1 and argued similarly
Claim Rejections under 35 U.S.C. 103- by virtue of dependency
Claim(s) 2-7, 10-15 and 17-20
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure from IP.com
Mutual information in the Tangled Nature Model, Arxiv 2009
Tangle Machines, Arxiv 2014
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BRUCE I EBERSMAN whose telephone number is (571)270-3442. The examiner can normally be reached 8:00 am - 5:00 pm Monday-Friday.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Michael W Anderson can be reached at 571-270-0508. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/BRUCE I EBERSMAN/Primary Examiner, Art Unit 3693
Prosecution Insights