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 .
Priority
Should applicant desire to obtain the benefit of foreign priority under 35 U.S.C. 119(a)-(d) prior to declaration of an interference, a certified English translation of the foreign application must be submitted in reply to this action. 37 CFR 41.154(b) and 41.202(e).
Failure to provide a certified translation may result in no benefit being accorded for the non-English application.
Specification
Applicant is reminded of the proper language and format for an abstract of the disclosure.
The language should be clear and concise and should not repeat information given in the title. It should avoid using phrases which can be implied, such as, “The disclosure concerns,” “The disclosure defined by this invention,” “The disclosure describes,” etc.
The abstract of the disclosure is objected to because it contains language that repeats information given in the title and uses phrases which can be implied. “A system for collaboration and optimization of edge machines based on federated learning is provided.” A corrected abstract of the disclosure is required and must be presented on a separate sheet, apart from any other text. See MPEP § 608.01(b).
The use of the term, “TensorFlow” in para. [0005], which is a trade name or a mark used in commerce, has been noted in this application. The term should be accompanied by the generic terminology; furthermore the term should be capitalized wherever it appears or, where appropriate, include a proper symbol indicating use in commerce such as ™, SM , or ® following the term.
Although the use of trade names and marks used in commerce (i.e., trademarks, service marks, certification marks, and collective marks) are permissible in patent applications, the proprietary nature of the marks should be respected and every effort made to prevent their use in any manner which might adversely affect their validity as commercial marks.
Claim Objections
Claims 1-10 are objected to because of the following informalities:
In claim 1, “a model parameter assignment unit, configured to” should read, “a model parameter assignment unit configured to”.
In claim 1, “the model training and optimization units…and configured to train” should read, “the model training and optimization units…are configured to train”.
Claims 2-10 are objected to for inheriting the deficiencies of their respective base claims.
In claim 2, “scenario feature model optimizing units, arranged in the
T
i
specific edge machines, and configured to”, should read, “scenario feature model optimizing units, arranged in the
T
i
specific edge machines, are configured to”.
Claim 3 is objected to for inheriting the deficiencies of claim 2.
In claim 3, “shorter than predetermined value” should read, “shorter than a predetermined value”.
In claim 6, “a machine selection unit, configured to” should read, “a machine selection unit configured to”.
In claim 6, “a task model parameter assignment unit, configured to” should read, “a task model parameter assignment unit configured to”.
In claim 6, “the task model training and optimizing units…and configured to train” should read, “the task model training and optimizing units…are configured to train”.
In claim 10, “a data acquisition module, configured to” should read, “a data acquisition module configured to”.
In claim 10, “a storage unit, configured to” should read, “a storage unit configured to”.
In claim 10, “an other part” should read, “another part”.
Appropriate correction is required.
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitations are in claims 1-2, 6-7, and 10:
Claim 1, a model parameter assignment unit, configured to
“assign initial parameters for federated learning to the
M
i
edge machines in the i-th federated learning system”
“receive intermediate model parameters transmitted by model training and optimizing units”
“aggregate and update the received intermediate model parameters to obtain new model parameters”
Claim 1, “the model and optimizing units, arranged in the
M
i
edge machines respectively, [are] configured to”
“train, on the basis of the initial parameters assigned by the model parameter assignment unit and respective operating data, local operating models”
“transmit the intermediate model parameters obtained after training the model parameter assignment unit”
“obtain a new system collaborative operating model of the i-th federated learning system according to the new model parameters”
Claim 2, “scenario feature model optimizing units, arranged in the
T
i
specific edge machines, [are] configured to carry out, on the basis of the system collaborative operating model and working scenario features of the
T
i
specific edge machines, model optimization”
Claim 6, “a machine selection unit, configured to select edge machines with performance scores of executing a target task higher than a predetermined value in each of the R federated learning systems to obtain a task training alliance”
Claim 6, “a task model parameter assignment unit, configured to”
“assign task initial parameters to the edge machines in the task training alliance”
“receive task model intermediate parameters transmitted by the task model training and optimizing units”
“aggregate and update the received task model intermediate parameters to obtain new task model parameters”
Claim 6, “the task model training and optimizing units, arranged in the edge machines in the task training alliance respectively, [are] configured to”
“train, on the basis of the task initial parameters assigned by the task model parameter assignment unit and respective operating data, local operating models for the target task”
“encrypt the task model intermediate parameters obtained after training”
”transmit the encrypted task model intermediate parameters to the task model parameter assignment unit”
“obtain a system collaborative execution task model of the task training alliance according to the new task model parameters”
Claim 7, “wherein the model parameter assignment unit is further configured for recording and making statistics on activity data in the federated learning systems”
Claim 10, “a data acquisition module, configured to acquire an image, a movement track, operating data and environment responding data”
Claim 10, “a storage unit, configured to store the operating data for model training”
Claim 10, “a computing unit, of which one part is configured to execute a predetermined working task and an other part is configured to execute a task of the federating learning”
Claim 10, “a communication module, which supports wired communication and wireless communication”
The examiner is interpreting the claimed functions to be implemented on a generic processor or computer because the applicant’s disclosure does not provide details about how they are implemented or how they are function aside from merely reciting the claim language.
Because this/these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 1-10 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claims contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
Regarding Claims 1-10:
As described above, the disclosure does not provide adequate structure to perform the claimed functions of:
Claim 1, a model parameter assignment unit, configured to
“assign initial parameters for federated learning to the
M
i
edge machines in the i-th federated learning system”
“receive intermediate model parameters transmitted by model training and optimizing units”
“aggregate and update the received intermediate model parameters to obtain new model parameters”
Claim 1, “the model and optimizing units, arranged in the
M
i
edge machines respectively, [are] configured to”
“train, on the basis of the initial parameters assigned by the model parameter assignment unit and respective operating data, local operating models”
“transmit the intermediate model parameters obtained after training the model parameter assignment unit”
“obtain a new system collaborative operating model of the i-th federated learning system according to the new model parameters”
Claims 2-10 are rejected for inheriting the deficiencies of claim 1.
Claim 2, “scenario feature model optimizing units, arranged in the
T
i
specific edge machines, [are] configured to carry out, on the basis of the system collaborative operating model and working scenario features of the
T
i
specific edge machines, model optimization”
Claim 3 is rejected for inheriting the deficiencies of claim 2.
Claim 6, “a machine selection unit, configured to select edge machines with performance scores of executing a target task higher than a predetermined value in each of the R federated learning systems to obtain a task training alliance”
Claim 6, “a task model parameter assignment unit, configured to”
“assign task initial parameters to the edge machines in the task training alliance”
“receive task model intermediate parameters transmitted by the task model training and optimizing units”
“aggregate and update the received task model intermediate parameters to obtain new task model parameters”
Claim 6, “the task model training and optimizing units, arranged in the edge machines in the task training alliance respectively, [are] configured to”
“train, on the basis of the task initial parameters assigned by the task model parameter assignment unit and respective operating data, local operating models for the target task”
“encrypt the task model intermediate parameters obtained after training”
”transmit the encrypted task model intermediate parameters to the task model parameter assignment unit”
“obtain a system collaborative execution task model of the task training alliance according to the new task model parameters”
Claim 7, “wherein the model parameter assignment unit is further configured for recording and making statistics on activity data in the federated learning systems”
Claim 10, “a data acquisition module, configured to acquire an image, a movement track, operating data and environment responding data”
Claim 10, “a storage unit, configured to store the operating data for model training”
Claim 10, “a computing unit, of which one part is configured to execute a predetermined working task and an other part is configured to execute a task of the federating learning”
Claim 10, “a communication module, which supports wired communication and wireless communication”
The specification merely recites the claim language and there are no algorithms disclosed for performing the claimed functions. Therefore, the specification does not demonstrate that the applicant has made an invention that achieves the claimed functions because the invention is not described with sufficient detail that one of ordinary skill in the art can reasonably conclude that the inventor had possession of the claimed invention.
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-10 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Regarding Claims 1-10:
Claims 1-2, 6-7, and 10 include claim limitations that invoke 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. However, the written description fails to disclose the corresponding structure, material, or acts for performing the entire claimed function and to clearly link the structure, material, or acts to the function. The claim limitations:
Claim 1, a model parameter assignment unit, configured to
“assign initial parameters for federated learning to the
M
i
edge machines in the i-th federated learning system”
“receive intermediate model parameters transmitted by model training and optimizing units”
“aggregate and update the received intermediate model parameters to obtain new model parameters”
Claim 1, “the model and optimizing units, arranged in the
M
i
edge machines respectively, [are] configured to”
“train, on the basis of the initial parameters assigned by the model parameter assignment unit and respective operating data, local operating models”
“transmit the intermediate model parameters obtained after training the model parameter assignment unit”
“obtain a new system collaborative operating model of the i-th federated learning system according to the new model parameters”
Claim 2, “scenario feature model optimizing units, arranged in the
T
i
specific edge machines, [are] configured to carry out, on the basis of the system collaborative operating model and working scenario features of the
T
i
specific edge machines, model optimization”
Claim 6, “a machine selection unit, configured to select edge machines with performance scores of executing a target task higher than a predetermined value in each of the R federated learning systems to obtain a task training alliance”
Claim 6, “a task model parameter assignment unit, configured to”
“assign task initial parameters to the edge machines in the task training alliance”
“receive task model intermediate parameters transmitted by the task model training and optimizing units”
“aggregate and update the received task model intermediate parameters to obtain new task model parameters”
Claim 6, “the task model training and optimizing units, arranged in the edge machines in the task training alliance respectively, [are] configured to”
“train, on the basis of the task initial parameters assigned by the task model parameter assignment unit and respective operating data, local operating models for the target task”
“encrypt the task model intermediate parameters obtained after training”
”transmit the encrypted task model intermediate parameters to the task model parameter assignment unit”
“obtain a system collaborative execution task model of the task training alliance according to the new task model parameters”
Claim 7, “wherein the model parameter assignment unit is further configured for recording and making statistics on activity data in the federated learning systems”
Claim 10, “a data acquisition module, configured to acquire an image, a movement track, operating data and environment responding data”
Claim 10, “a storage unit, configured to store the operating data for model training”
Claim 10, “a computing unit, of which one part is configured to execute a predetermined working task and an other part is configured to execute a task of the federating learning”
Claim 10, “a communication module, which supports wired communication and wireless communication”
The specification is devoid of adequate structure to perform the claimed functions. The specification merely recites the claim language. There is no disclosure of any particular structure, either explicitly or inherently, to perform the claimed functions. As would be recognized by those of ordinary skill in the art, the claimed functions can be performed on a generically recited computer or processor. The specification does not provide sufficient details such that one of ordinary skill in the art would understand which structure or structures perform(s) the claimed functions. Therefore, claims 1-10 are indefinite and are rejected under 35 U.S.C. 112(b) or pre-AIA 35 U.S.C. 112, second paragraph. Additionally, claims 2-10 and 3 are rejected for inheriting the deficiencies of their base claim.
Applicant may:
(a) Amend the claim so that the claim limitation will no longer be interpreted as a limitation under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph;
(b) Amend the written description of the specification such that it expressly recites what structure, material, or acts perform the entire claimed function, without introducing any new matter (35 U.S.C. 132(a)); or
(c) Amend the written description of the specification such that it clearly links the structure, material, or acts disclosed therein to the function recited in the claim, without introducing any new matter (35 U.S.C. 132(a)).
If applicant is of the opinion that the written description of the specification already implicitly or inherently discloses the corresponding structure, material, or acts and clearly links them to the function so that one of ordinary skill in the art would recognize what structure, material, or acts perform the claimed function, applicant should clarify the record by either:
(a) Amending the written description of the specification such that it expressly recites the corresponding structure, material, or acts for performing the claimed function and clearly links or associates the structure, material, or acts to the claimed function, without introducing any new matter (35 U.S.C. 132(a)); or
(b) Stating on the record what the corresponding structure, material, or acts, which are implicitly or inherently set forth in the written description of the specification, perform the claimed function. For more information, see 37 CFR 1.75(d) and MPEP §§ 608.01(o) and 2181.
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 1 and 9 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Xie (“Multi-Center Federated Learning to Cluster Clients with non-IID data”).
Regarding Claim 1:
Xie discloses:
A system for collaboration and optimization of edge machines based on federated learning, comprising:
Xie, pg. 37, “Federated Learning (FL)…is a decentralized machine learning framework that learns models collaboratively using the training data distributed on remote devices to boost communication efficiency.”
Xie discloses federated learning which refers to a framework that collaboratively and efficiently works with remote devices [collaboration and optimization of edge machines].
R federated learning systems, wherein R ≥ 1, an i-th federated learning system in the R federated learning systems comprises Mi edge machines with uneven operating experience distribution, Mi ≥ 2, i = 1, 2,…, R
Xie, pg. 42, “we propose a novel model aggregation method with multiple centers, each associating with a global model
W
~
(
k
)
updated by aggregating a cluster of user’s models with nearly IID data. In particular, all the local models will be grouped to
K
clusters, denoted as
C
1
,
⋯
,
C
K
, each covering a subset of local models with parameters
{
W
j
}
j
=
1
m
k
. An intuitive comparison between the vanilla FL and our multi-center FL is illustrated in Fig. 3.1.”
Pg. 17, “The insight to FL is such: under these setups, the performances of the training process may vary significantly according to the degree of unbalanced local data samples”
PNG
media_image1.png
573
907
media_image1.png
Greyscale
On pg. 42, Xie discloses multi-center federated learning which is depicted in Fig. 3.1 (from pg. 41) [R federated learning systems, wherein R ≥ 1]. Xie further details that each
K
cluster of the multiple clusters [an i-th federated learning system in the R federated learning system] is an aggregate of local models
C
1
,
⋯
,
C
K
[comprises Mi edge machines…Mi ≥ 2, i = 1, 2,…, R]. Lastly, referring to pg. 17, Xie discloses that in federated learning, the performance of training varies due to unbalanced local data samples [edge machines with uneven operating experience distribution].
a model parameter assignment unit, configured to
Xie, pg. 22, “To begin with, workers send their loss to centralised server, center machine estimates the cluster identities of each worker machine by running k-means on the collection of workers local models. Then with the cluster identity estimations, the center machine runs any federated learning algorithm…”
Xie discloses a centralized server (also described as a center machine or cluster center) [a model parameter assignment unit]. The center is interpreted as the claimed model parameter assignment unit because both perform the claimed functions.
assign initial parameters for federated learning to the Mi edge machines in the i-th federated learning system
Xie, pg. 45, “to update the local models, the global model’s parameters
W
~
(
k
)
are sent to each device in cluster
k
to update its local model, and then we can fine-tune the local model’s parameters
W
i
using a supervised learning algorithm on its own private training data…to update the local model, we need to fine-tune the local model by implementing Algorithm 2.”
Pg. 46,
PNG
media_image2.png
182
348
media_image2.png
Greyscale
On pg. 45, Xie discloses using the center cluster’s global model’s parameters [assign initial parameters for federated learning] to initialize the local models in the cluster [to the Mi edge machines in the i-th federated learning system]. This is further depicted in the Initialization step of Algorithm 2.
receive intermediate model parameters transmitted by model training and optimizing units
In Algorithm 2, Xie discloses using the local models [training and optimizing units] to fine-tuned parameters [intermediate model parameters transmitted] to update the cluster center. The last line of Algorithm 2, cited above, further discloses that the local parameters are transmitted to the server/cluster center. The edge devices with local models correspond to training and optimizing units because both refer to the local models perform the training and loss optimization locally on the edge devices.
aggregate and update the received intermediate model parameters to obtain new model parameters
Xie, pg. 45, “for the M-Step, we update the cluster center
W
~
(
k
)
according to the
W
i
and
r
i
(
k
)
”
Xie discloses using each of the local
W
i
[aggregate and update the received intermediate model parameters] to update the cluster center’s parameters,
W
~
(
k
)
[obtain new model parameters].
the model training and optimizing units, arranged in the Mi edge machines respectively, and configured to
Xie, pg. 45, “to update the local model, we need to fine-tune the local model by implementing Algorithm 2.”
Xie discloses using Algorithm 2 to train and optimize the local models [the model training optimizing units]. Furthermore, the algorithm is implemented by the local models [arranged in the Mi edge machines respectively]. The algorithm is interpreted as the claimed model training and optimizing units because both perform the claimed functions.
train, on the basis of the initial parameters assigned by the model parameter assignment unit and respective operating data, local operating models
Pg. 46,
PNG
media_image2.png
182
348
media_image2.png
Greyscale
Xie discloses training local models [train…local operating models] using the model parameters from the server [on the basis of the initial parameters assigned by the model parameter assignment unit] and the local training data [respective operating data].
transmit the intermediate model parameters obtained after training to the model parameter assignment unit
As seen above in Algorithm 2, the locally updated parameters
W
i
[the intermediate model parameters obtained after training] are returned to the server [transmit…to the model parameter assignment unit].
obtain a system collaborative operating model of the i-th federated learning system according to the new model parameters
Xie, pg. 44, “we sequentially conduct: 1) E-step – updating cluster assignment
r
i
(
k
)
with fixed
W
i
, 2) M-step – updating cluster centers
W
~
(
k
)
, and 3) updating local models by providing new initialization
W
~
(
k
)
.”
Pg. 45, “Lastly, we repeat the three stochastic updating steps above until convergence.”
Xie discloses repeating the above steps. Therefore, after updating the global model [a system collaborative operating model], the updated global model parameters are to be sent to the local models [obtain a system collaborative operating model…according to the new model parameters] for each cluster [of the i-th federated learning system].
wherein the local operating models are models in response to different operating environments
Xie, pg. 4, “FL builds a joint model using the data located at different sites, where each party contributes some data to train the model. The devices can be owned by different individuals or organizations, and can be of different types (e.g., smartphones, sensors, vehicles, etc.).”
Xie discloses that each of the local models are from devices that can be from different individuals or organizations [models in response to different operating environments].
Regarding Claim 9:
As discussed above, Xie teaches [the] system according to claim 1, and Xie further discloses:
wherein a cloud server or an edge server capable of communicating with the Mi edge machines serves as the model parameter assignment unit
Xie, pg. 22, “To begin with, workers send their loss to centralised server, center machine estimates the cluster identities of each worker machine by running k-means on the collection of workers local models. Then with the cluster identity estimations, the center machine runs any federated learning algorithm…”
Xie discloses a centralized server (also described as a center machine or cluster center) capable of communicating with workers or edge machines [an edge server capable of communicating with the Mi edge machines serves as the model parameter assignment unit].
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Xie in view of Pang et al. (“Realizing the Heterogeneity: A Self-Organized Federated Learning Framework for IoT”), hereinafter Pang.
Regarding Claim 2:
As discussed above, Xie teaches [the] system according to claim 1, but does not explicitly disclose:
wherein the Mi edge machines comprise Ti specific edge machines with operating experience not meeting predetermined requirements,
1
≤
T
i
<
M
i
and the system further comprises: scenario feature model optimizing units, arranged in the Ti specific edge machines, and configured to carry out, on the basis of the system collaborative operating model and working scenario features of the Ti specific edge machines, model optimization
to increase single machine intelligence and improve capabilities of the Ti specific edge machines to respond to environments, in which the Ti specific edge machines are located, and to execute tasks
However, in the same field, analogous art Pang teaches:
wherein the Mi edge machines comprise Ti specific edge machines with operating experience not meeting predetermined requirements,
1
≤
T
i
<
M
i
Pang, pg. 5, col. 2, “we regard the FL [federated learning] aggregation process as a self-organizing system, in which each client
c
i
can recognize how the collaboration impacts their income and the collaboration plan
p
i
can always toward better performance achievements
G
(
p
i
)
. To achieve such a self-organizing-based FL aggregation process, we introduce a feedback mechanism to collect every client’s evaluation instead of the direct interactions among clients. This feedback mechanism can effectively reflects how the collaboration plan
p
i
satisfy the client’s expectation on
g
p
i
(
c
i
)
without disturbing the privacy settings of FL principles.”
Pang discloses evaluating client devices within a federated learning frame [Mi edge machines] using feedback to determine which client devices are not satisfying performance [Ti specific edge machines with operating experience not meeting predetermined requirements,
1
≤
T
i
<
M
i
].
and the system further comprises: scenario feature model optimizing units, arranged in the Ti specific edge machines, and configured to carry out, on the basis of the system collaborative operating model and working scenario features of the Ti specific edge machines, model optimization
Pang, pg. 5, col. 2, “To be specific, we assume the clients submit their feedback to the central server to rate the current collaboration plan
p
i
in each iteration. After that, the updated collaboration plan
p
i
+
1
based on the new rating feedback will help the central server adjust the current collaboration plan. The process is repeated until the performance improvements
G
becomes stable. In this process, the rating feedback refers to individual performance changes of the current collaboration plan over the previous one.”
Pg. 1, col. 1, “To be specific, an FL central server analyses the benefits of different collaboration by capturing the intricate patterns in heterogeneous clients based on rating feedback and then updates clients’ weights iteratively, until it establishes a coalition of clients with quasioptimal performance.”
Pang discloses that clients with unsatisfactory performance use rating feedback to update their weights until the clients are able to achieve stable performance [carry out, on the basis of the system collaborative operating model and working scenario features of the Ti specific edge machines, model optimization].
to increase single machine intelligence and improve capabilities of the Ti specific edge machines to respond to environments, in which the Ti specific edge machines are located, and to execute tasks
As cited above, Pang discloses updating clients’ weights iteratively until quasioptimal performance is established. Therefore, the clients’ performance and intelligence are improved when performing execution of tasks.
Xie, Pang, and the instant application are analogous art because they are all directed to federated learning.
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Xie with Pang to improve the performance of individual client devices. “We proposed a novel FL aggregation method. The work first adopts the concept of the self-organizing process and implements such an idea by RL. Our solution has several advantages. First, it can identify the heterogeneity level automatically by testing different collaboration plans during Fig. 7. Cumulative reward versus the number of episodes. aggregation, and it can also find the consensus clients to form a coalition. Second, an RL-based solution is proposed for weight allocation considering the coalition. Extensive simulations with real data sets reveal that the proposed framework significantly improves the performance of the majority of clients” (Pang, pg. 10)
Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over Xie in view of Pang, and further in view of Sprague et al. (US 20200334524), hereinafter Sprague.
Regarding Claim 3:
As discussed above, Xie in view of Pang teach [the] system according to claim 2, but do not explicitly disclose:
wherein the operating experience not meeting the predetermined requirements comprises one of:
a number of operating scenarios experienced being lower than a predetermined value;
a quantity of operating data being less than a predetermined value; or
operating duration being shorter than predetermined value
However, in the same field, analogous art Sprague teaches:
wherein the operating experience not meeting the predetermined requirements comprises one of:
a number of operating scenarios experienced being lower than a predetermined value;
a quantity of operating data being less than a predetermined value; or
Sprague, [0030], “a threshold (τ) is set to the number of data points that should be processed by the worker device before it can send a parameter vector to the server.”
Sprague discloses using a threshold [less than a predetermined value] to determine the number of data points that need to be processed [a quantity of operating data].
operating duration being shorter than predetermined value
Xie, Pang, Sprague, and the instant application are analogous art because they are all directed to decentralized learning.
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Xie and Pang with Sprague to identify when an edge device should process more data in order to retrain the model and improve its performance. “The device processor is configured to train the model using a threshold quantity of the data instances of the plurality of data instances. The device processor is configured to over sample or under sample the plurality of data instances to equal the threshold quantity. The device processor is further configured to transmit a parameter vector of the trained model to the parameter server and receive in response, an updated central parameter vector from the parameter server derived from the model; the device processor further configured to retrain the model using the updated central parameter vector. The at least one sensor acquires different data instances than other sensors of the other devices that are training respective models.” (Sprague, [0005]).
Claim 4-5 are rejected under 35 U.S.C. 103 as being unpatentable over Xie in view of Brown et al. (US 20100262829), hereinafter Brown, and further in view of Sirdey et al. (US 20200394518), hereinafter Sirdey.
Regarding Claim 4:
As discussed above, Xie teaches [the] system according to claim 1, and but does not explicitly disclose:
wherein when the Mi edge machines in the i-th federated learning system are organizations with visible data privacy, the intermediate model parameters are transmitted without encryption
when the Mi edge machines in the i-th federated learning system are organizations with invisible data privacy, the intermediate model parameters need to be transmitted with encryption
In the same field, analogous art Brown teaches:
…organizations with visible data privacy…the…parameters are transmitted without encryption
Brown, [0161], “In one variant embodiment, the security parameters are transmitted in unencrypted form. The present inventors recognized that where the one or more security parameters to be transmitted comprises one or more public keys, there would be no need to keep the data secret (and therefore encrypt it) because it is already "public" information (i.e. anyone may have access to it).”
Brown discloses a system where parameters are sent in unencrypted form [parameters are transmitted without encryption] because there’s no need to keep the data secret [organizations with invisible data privacy].
Xie, Brown, and the instant application are analogous art because they are all directed to data transmission over a network.
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Xie with Brown to leave data unencrypted in order to avoid unnecessary encryption when information can be readily accessed. “The present inventors recognized that where the one or more security parameters to be transmitted comprises one or more public keys, there would be no need to keep the data secret (and therefore encrypt it) because it is already "public" information (i.e. anyone may have access to it)” (Brown, [0161]).
However, in the same field, analogous art Sirdey teaches:
when the Mi edge machines in the i-th federated learning system are organizations with invisible data privacy, the intermediate model parameters need to be transmitted with encryption
Sirdey, [0042], “It will be noted that in the federative learning process described above, the training data are never encrypted and/or transmitted homomorphically to the platform, only the parameters of the current model are.”
[0001], “This invention relates…particularly to the field of collaborative learning of an artificial neural network. It also relates to the privacy-preserving computation field.”
Sirdey discloses a federative learning process, where the current model’s parameters [the intermediate model parameters] are transmitted homomorphically to the platform [transmitted with encryption]. As stated in para. 1, the collaborative or federative learning process is privacy preserving [the Mi edge machines in the i-th federated learning system are organizations with invisible data privacy].
Xie, Brown, Sirdey, and the instant application are analogous art because they are all directed to data transmission over a network.
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Xie and Brown with Sirdey to perform encryption in order to optimize transmission rates as well as preserve privacy. “It will be noted that in the federative learning process described above, the training data are never encrypted and/or transmitted homomorphically to the platform, only the parameters of the current model are. The parameters represent a much smaller quantity of data then the training data, a fortiori after encryption in the homomorphic domain. Thus, the learning process does not require very high transmission rates” (Sirdey, [0042]).
Regarding Claim 5:
As discussed above, Xie in view of Brown, and further in view of Sirdey teach [the] system according to claim 4, and Sirdey further discloses:
wherein the encryption comprises homomorphic encryption
Sirdey, [0042], “The parameters represent a much smaller quantity of data then the training data, a fortiori after encryption in the homomorphic domain.”
the homomorphic encryption comprises fully homomorphic encryption
Sirdey, [0060], “It should be noted that combination operations more complex than those in (1) could be envisaged without going outside the scope of the present invention. In this case, we could use FHE (Full Homomorphic Encryption)…”
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Xie, Brown, and Sirdey, further with Sirdey to perform fully homomorphic encryption in order to optimize transmission rates as well as preserve privacy. “It will be noted that in the federative learning process described above, the training data are never encrypted and/or transmitted homomorphically to the platform, only the parameters of the current model are. The parameters represent a much smaller quantity of data then the training data, a fortiori after encryption in the homomorphic domain. Thus, the learning process does not require very high transmission rates” (Sirdey, [0042]).
Claim 6 is rejected under 35 U.S.C. 103 as being unpatentable over Xie in view of AbdulRahman et al. (US "FedMCCS: Multicriteria Client Selection Model for Optimal IoT Federated Learning"), hereinafter Abdul, and further in view of Sirdey.
Regarding Claim 6:
As discussed above, Xie teaches [the] system according to claim 1, and but does not explicitly disclose:
a machine selection unit, configured to select edge machines with performance scores of executing a target task higher than a predetermined value in each of the R federated learning systems to obtain a task training alliance
Abdul, pg. 5, col. 1, “4) Multicriteria-Based Client Selection Engine: It is the main module in our system. It allows to select the maximum number of available and performant clients (colored devices in the figure) per round in order to aggregate their updates. The proposed selection algorithm employs multiple criteria (time, CPU, memory, and energy) to achieve the maximization goal.
7) Machine Learning Engine: It is responsible of performing the training task using the local data stored on the device.”
Abdul discloses a Multicriteria-Based Client Selection Engine [a machine selection unit] to select clients [edge machines] based on maximizing criteria such as time, CPU, memory, and energy [with performance scores…higher than a predetermined value in each of the R federated learning systems to obtain a task training alliance]. The models on the clients are responsible for performing the task [executing a target task].
a task model parameter assignment unit, configured to
assign task initial parameters to edge machines in the task training alliance
Abdul, pg. 5, col. 2, Protocol 3, “Protocol 3 FL [federated learning] With Multicriteria Client Selection…5: Distribution : The server disseminates the global model parameters to the selected clients”
Abdul discloses sending global model parameters [assign task initial parameters] to the selected clients [edge machines in the task training alliance]
receive task model intermediate parameters transmitted by task model training and optimizing units
Abdul, pg. 5, col. 2, Protocol 3, “6: Update and Upload in Protocol 1.”
Pg. 3, col. 2, Protocol 1, “4: Update and Upload : Selected clients use their local data to update the shared model and upload the new model parameters to the server.”
Abdul discloses receiving the new model parameters from select clients [receive task model intermediate parameters transmitted by task model training and optimizing units].
aggregate and update the received task model intermediate parameters to obtain new task model parameters
Abdul, pg. 5, col. 2, Protocol 3, “7: Aggregation : The server averages the parameters, when more than 70% of the requested updates are received.”
Abdul discloses aggregating the received parameters [aggregate and update the received task model intermediate parameters to obtain new task model parameters].
the task model training and optimizing units, arranged in the edge machines in the task training alliance respectively, and configured to
train, on the basis of the task initial parameters assigned by the task model parameter assignment unit and respective operating data, local operating models for the target task
Abdul, pg. 5, col. 2, Protocol 3, “5: Distribution : The server disseminates the global model parameters to the selected clients
6: Update and Upload in Protocol 1.”
Pg. 3, col. 2, Protocol 1, “4: Update and Upload : Selected clients use their local data to update the shared model and upload the new model parameters to the server.”
In Protocol 3 Abdul discloses updating the local model of the selected clients [train…local operating models for the target task] using the global model parameters and the local data [on the basis of the task initial parameters assigned by the task model parameter assignment unit and respective operating data].
obtain a system collaborative execution task model of the task training alliance according to the new task model parameters
Abdul, pg. 5, col. 2, “7: Aggregation : The server averages the parameters, when more than 70% of the requested updates are received.
8: All steps but Initialization are iterated as in Protocol 2.”
Pg. 4, col. 2, Protocol 2, “7: Steps 2 till 5 are repeated until achieving a desired performance of the model or when the final deadline is met.”
Abdul discloses aggregating the parameters [according to the new task model parameters] and iterating until a desired performance performing steps iteratively until a model with a desired performance is achieved [obtain a system collaborative execution task model of the task training alliance].
wherein the local operating models for the target task are models for executing the target task in different operating environments
Abdul, pg. 4, col. 1, “Existing protocols of FL, especially when applied in the IoT environment [3], [31], [32], engender many problems in the local training phase. The clients are very heterogeneous in nature. They have different communication and computation resources, variant amount of generated data, and different time zones.”
Abdul discloses that the clients [the local operating models for the target task are models for executing the target task] in the IoT environment are varying in nature [different operating environments].
Xie, Abdul, and the instant application are analogous art because they are all directed to federated learning.
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Xie with Abdul to select specific clients that are best performing for performing a task. “Multicriteria-Based Client Selection Engine: It is the main module in our system. It allows to select the maximum number of available and performant clients (colored devices in the figure) per round in order to aggregate their updates” (Abdul, pg. 5, col. 1).
Xie in view of Abdul do not explicitly disclose:
encrypt the task model intermediate parameters to the task model parameter assignment unit
However, in the same field, analogous art Sirdey teaches:
encrypt the task model intermediate parameters to the task model parameter assignment unit
Sirdey, [0034], “According to one variant, each data supplier encrypts the parameters of his partial model using a secret key, for example a symmetric key of a stream encryption, instead of encrypting them homomorphically. The data suppliers then send the parameters encrypted using their own secret keys to the platform.
Sirdey discloses encrypting their data supplier’s parameters [the task model intermediate parameters] to the platform [the task model parameter assignment unit].
Xie, Abdul, Sirdey, and the instant application are analogous art because they are all directed to federated learning.
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Xie and Abdul with Sirdey to perform encryption in order to optimize transmission rates as well as preserve privacy. “It will be noted that in the federative learning process described above, the training data are never encrypted and/or transmitted homomorphically to the platform, only the parameters of the current model are. The parameters represent a much smaller quantity of data then the training data, a fortiori after encryption in the homomorphic domain. Thus, the learning process does not require very high transmission rates” (Sirdey, [0042]).
Claim 7 is rejected under 35 U.S.C. 103 as being unpatentable over Xie in view of Wang et al. (US 20220101195), hereinafter Wang, and further in view of Choudhary et al. (US 11593634), hereinafter Choudhary.
Regarding Claim 7:
As discussed above, Xie teaches [the] system according to claim 1, but does not explicitly disclose:
wherein the model parameter unit is further configured for recording and making statistics on activity data in federated learning systems,
wherein recording and making statistics on activity data in federated learning systems comprise:
a number of edge machines participating in computation
a number of model transfers
transmission and convergence determination of the updated model parameters
However, in the same field, analogous art Wang teaches:
wherein the model parameter unit is further configured for recording and making statistics on activity data in federated learning systems,
Wang, [0043], “the host device 110 establishes an associated program according to information of the selected client device, such information includes the identifiers ID1-ID3 of the client devices”
Wang discloses a program on the host device [the model parameter unit] that records information of the client device [recording and making statistics on activity data in federated learning systems].
wherein recording and making statistics on activity data in federated learning systems comprise:
a number of edge machines participating in computation
Wang, [0043], “In some embodiments, the identifiers IDs are configured to record identities of the selected client devices”
Wang discloses recording identities of the selected client devices [a number of edge machines participating in computation].
transmission and convergence determination of the updated model parameters
Wang, [0043], “the round numbers are configured to record the times that the selected client devices conducting trainings and generating the training results”
Wang discloses recording the times that the client devices conducting trainings and generating the training results [transmission and convergence determination of the updated model parameters].
Xie, Wang, and the instant application are analogous art because they are all directed to federated learning.
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Xie with Wang to record information regarding each client device in order to achieve better model performance. “Based on above, the machine learning system 100 and the machine learning method 200 provided by the present disclosure, by controlling a difference of the round numbers between each of the client devices, render each of the client devices may generate training results asynchronously under the difference of the round numbers not greater than the threshold value, so as to reduce the time waste caused by each client device waiting for each other, and alleviate the problem that the accuracy of the host model is affected by excessive differences of the round numbers between each client device” (Wang, [0048]).
Xie in view of Wang do not explicitly disclose:
wherein recording and making statistics on activity data in federated learning systems comprise:
a number of model transfers
However, in the same field, analogous art Choudhary teaches:
wherein recording and making statistics on activity data in federated learning systems comprise:
a number of model transfers
Choudhary, [19], “the asynchronous training system can tally or track the number of training iterations in which a client device sends modified parameter indicators.”
Choudhary discloses tracking the number of training iterations which includes sending parameters.
Xie, Wang, Choudhary, and the instant application are analogous art because they are all directed to federated learning.
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Xie and Wang with Choudhary to record the number of model transfers in order to optimize communications between devices. “As also noted above, in some embodiments, the asynchronous training system 106 tallies or tracks the number of training iterations in which a client device sends modified parameter indicators and applies a bounded delay approach to include a broad range of client devices without unduly slowing the training process” (Choudhary, [82]).
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Xie in view of Vestias et al. (“Client-Edge-Cloud Hierarchical Federated Learning”), hereinafter Vestias.
Regarding Claim :
As discussed above, Xie teaches [the] system according to claim 1, but does not explicitly disclose:
wherein an edge machine with a computing capability and storage capability meeting predetermined requirements in the Mi edge machines serves as the model parameter assignment unit
Vestias, pg. 19, “all devices update their parameters to the central edge coordinator…”
Vestias discloses the central edge coordinator [an edge machine… in the Mi edge machines serves as the model parameter assignment unit]. Since Vestias discloses that the central edge coordinator is able to communicate to all devices, it is construed as disclosing an edge machine with computing capability and storage capability meeting predetermined requirements to serve as the model parameter assignment unit.
Xie, Vestias, and the instant application are analogous art because they are all directed to machine learning over a network.
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Xie with Vestias to use an edge device as the model parameter assignment unit in order to increase energy efficiency. “Edge servers and end devices have much less computing and memory resources compared to a cloud data center [22]. Several research directions have been followed to overcome these limitations: (1) new architectures for edge servers integrated with cloud servers for an efficient computational coordination [23]; (2) new DL models with computing, memory, and energy concerns; (3) new hardware-oriented optimizations to improve performance and energy efficiency; (4) and new computing architectures oriented toward DL models” (Vestias, pg. 2).
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Xie in view of McCann et al. (US 9977425), hereinafter McCann.
Regarding Claim 10:
As discussed above, Xie teaches [the] system according to claim 1, and further discloses:
wherein each of the Mi edge machines in the i-th federated learning system in the R federated learning systems further comprises:
a storage unit, configured to store the operating data for model training
Xie, pg. 25, “In FL, since the data source and computing nodes are end users’ personal devices…”
Xie discloses the client devices are their own data source, and is construed as having their own storage to store the operating data for model training.
a computing unit, of which one part is configured to execute a predetermined working task
Xie, pg. 53, “In FL, each device-i has a private dataset… Each dataset Di will be used to train a local supervised learning model…It is built to solve a specific task, and all devices share the same model architecture.”
Xie discloses each device has a local model built to solve a specific task.
and an other part is configured to execute a task for federated learning
As cited above on Xie, pg. 25, it is disclosed that each client device has its own computing node for federated learning.
a communication module, which supports wired and wireless communication
Xie, pg. 37, “Communication is a significant challenge in federated networks [65]. It has commonly occurred that wireless and end-use devices operate on lower bandwidth than interlink between data centers and can be expensive and unstable. This has led to significant recent interest in solutions to communication cost reduction of federated learning.”
Xie discloses wireless communication and interlink [wired] communication for federated networks
wherein the wireless communication involves a 5G communication module
Xie, pg. 37, “The variance of devices in federated networks can be huge, as well as the geographic location of the device, the computing capacity, the storage, and the network bandwidth used (3G, 4G, 5G, 6G, optics) of each user may differ to a large degree.”
Xie discloses different network bandwidths used by federated networks including 5G.
However, Xie does not explicitly disclose:
a data acquisition module, configured to acquire an image, a movement track, operating data and environment responding data
In the same field, analogous art McCann teaches:
a data acquisition module, configured to acquire
an image,
McCann, [33] “In addition to measured sensor data, the data host/server also receives image data”
McCann discloses receiving image data.
a movement track,
McCann, [33] “The image data is taken by equipment at the additive manufacturing machine and may be in the form of video and/or still images.”
McCann discloses receiving video data, interpreted as a movement track.
operating data and
McCann, [14], “It may receive inspection and/or operational data”
environment responding data
McCann, [16], “edge gateway 240 may receive sensor output data from one or more additive manufacturing machines 210”
McCann discloses receiving sensor data, interpreted as environment responding data.
Xie, McCann, and the instant application are analogous art because they are all directed to communications over networks.
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Xie with McCann to transmit data efficiently. “As shown in FIG. 5B, the sensor data is received via the network 572 by the data host/server 270 (see FIG. 2). The data host/server 270, in turn, makes the sensor data available to the user interface device 275, of which there may be many connected to the data host/server 270. The sensor data may be decompressed by the data host/server 270 or by the user interface device 275, the latter approach having the advantage that the data may be sent from the data host/server 270 to a number of user interface devices 275 in compressed form (see FIG. 2)” (McCann, [28]).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to STEVEN PHUNG whose telephone number is (703) 756-1499. The examiner can normally be reached Monday-Thursday: 9:00AM-4:00PM ET.
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, KAMRAN AFSHAR can be reached at (571) 272-7796. 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.
/STEVEN PHUNG/Examiner, Art Unit 2125
/KAMRAN AFSHAR/Supervisory Patent Examiner, Art Unit 2125
Prosecution Insights