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
1. This Office Action is in response to the preliminary amendment filed on 06/23/2023.
Claims 5, 12, 14 and 15 have been amended.
Claims 16-20 have been added.
Claims 1-20 are pending.
Priority
2. Receipt is acknowledged of papers submitted under 35 U.S.C. 119(a)-(d), which papers have been placed of record in the file.
Information Disclosure Statement
3. The information disclosure statement (IDS) filed on 09/25/2023 complies with the provisions of M.P.E.P. 609. The examiner has considered it.
Claim Rejections - 35 USC § 101
4. 35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
5. Claims 1, 9, 14 and 15 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter.
Claims 1, 9, 14 and 15 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Independent claims 1, 9, 14 and 15 recite:
A quantization method for a weight of a neural network, wherein the neural network is implemented on the basis of a crossbar-enabled analog computing-in-memory system, and the method comprises:
acquiring a distribution characteristic of the weight; and
determining, according to the distribution characteristic of the weight, an initial quantization parameter for quantizing the weight to reduce a quantization error in quantizing the weight.
Step 2A Prong One:
The limitations acquiring a distribution characteristic of the weight, and determining, according to the distribution characteristic of the weight, an initial quantization parameter for quantizing the weight to reduce a quantization error in quantizing the weight that, under its broadest reasonable interpretation, covers performance of the limitation in the mind, but for the recitation of generic computer components. That is , other than reciting, “computer method’; nothing in the claim element precludes the step from practically being performed in a human mind. Note that the limitations are done by the generically recited computer components under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes’ grouping of abstract ideas (concepts performed in the human mind including an observation, evaluation, judgment, and opinion).
Step 2A Prong Two:
The judicial exception is not integrated into a practical application. Claims 1, 9, 14 and 15 recite the additional element, “to reduce a quantization error in quantizing the weight.” these limitation is a mere generic intended usage or consequence of using the distribution of the weight and an initial quantization parameter. The limitations amount to a data gathering step and a mere generic presentation of collected and analyzed data which is considered to be insignificant extra solution activity (see MPEP 2106.05(g)).
Step 2B:
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. The limitation “to reduce a quantization error in quantizing the weight”, are recognized by the courts as well-understood, routine , and conventional activities when they are claimed in a merely generic manner (see MPEP 2106.05(qd)/(II) (iv) transferring and/or displaying information, Versata Dev. Group Inc.
Dependent Claims 2-8, 10-13 and 16-20
These claims are eligible due to their integration into a practical application.
Claim Rejections - 35 USC § 112
6. 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.
7. Claims 4, 5, 6, 9, 14, 16-18 and 20 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 pre-AIA the applicant regards as the invention.
Claims 4, 5, 6, 9, 14, 16-18 and 20 are rejected due to duplicate claims.
Claim 16 and claim 17 are method claims and similar in scope to the method claim 4.
Claim 18 is a method claim and similar in scope to the method claim 5.
Claim 14 is an apparatus claim and similar in scope to the apparatus claim 9.
Claim 20 is an apparatus claim and similar in scope to the apparatus claim 6.
Examiner’s Note
8. A neural network and a crossbar-enabled-analog computing-in-memory system (According to Google): “A neural network can be implemented as a crossbar-enabled analog computing-in-memory (CACIM) system, which leverages memristor or resistive memory (ReRAM) crossbars to perform massive matrix-vector multiplications in parallel using physical laws. This architecture accelerates deep neural networks (DNNs) by storing synaptic weights directly in the analog conductances of memory devices at the crossbar crosspoints.”
A weight distribution (According to Google): “A distribution characteristic of weight defines how an object's or system's total mass is dispersed across its structure, components, or in a statistical set. It is a fundamental concept in engineering, safety, and data analysis.”
A distribution characteristic of the weight (According to the instant specification):
Paragraph 52: “taking the probability density distribution of the weight as the distribution characteristic of the weight is only exemplary” and “For example, the distribution characteristic of the weight may also include a cumulative probability density distribution of the weight”.
Liu et al, US 20200394522, [Liu: Paragraph 3 (“the neural network continuously revises weights and thresholds of the network to reduce an error function along a direction of negative gradient and approach an expected output”)] [Liu: Paragraph 7 (“obtaining an analyzing result of each type of the data to be quantized, in which the data to be quantized includes at least one type of neurons, weights, gradients, and biases of the neural network”, i.e., quantizing the weight using the initial quantization parameter)] [Liu: Paragraph 52 (“At the beginning of neural network training, the weight needs to be initialized randomly”)] [Liu: Paragraphs 55-58 (“The derivatives are called gradients, which are then used to calculate the gradients of the penultimate layer in the neural network. The process is repeated until the gradient corresponding to each weight in the neural network is obtained. Finally, the corresponding gradient is subtracted from each weight in the neural network, then the weight is updated once, to reduce errors” and “when quantized objects are weights of a whole neural network and the quantized weights are 8-bit fixed-point numbers, since a neural network usually contains millions of connections, almost all the space is occupied by weights that are connected with neurons”, i.e., “the gradients of the penultimate layer in the neural network” or “the quantized weight” is considered as ‘a distribution characteristic of the weight’ and the updated weight is considered as ‘an initial quantization parameter’)] [Liu: Paragraphs 56-57 (“the second stage is to perform the back propagation on a gradient to update weights in the trained neural network”, i.e., using the updated weight (‘an initial quantization parameter’) to train the neural network)] [Liu: Paragraph 58 (“when quantized objects are weights of a whole neural network and the quantized weights are 8-bit fixed-point numbers, since a neural network usually contains millions of connections, almost all the space is occupied by weights that are connected with neurons”, i.e., “the quantized weights” are considered as the ‘initial quantization parameters’)].
Lu et al, US 20210089922, [Lu: Abstract and paragraphs 4 and 53 (“quantizing a weight parameter of the deep neural network and/or an activation parameter of the deep neural network based on the quantization bit-width”)] [Lu: Paragraph 40 (“the error gradient of weights may be calculated over a small number of examples, so that the calculated gradient approximates the true error gradient. This approximation method may be referred to as stochastic gradient descent. Stochastic gradient descent may be repeated until the achievable error rate of the entire system has stopped decreasing or until the error rate has reached a target level.”)] [Lu: Paragraphs 39 and 74(“In lower layers, the gradient may depend on the value of the weights and on the computed error gradients of the higher layers. The weights may then be adjusted to reduce the error. This manner of adjusting the weights may be referred to as “back propagation” as it involves a “backward pass” through the neural network”, i.e., ‘backward propagation to obtain the updated weight’)] [Lu: Paragraph 74 (“In lower layers, the gradient may depend on the value of the weights and on the computed error gradients of the higher layers. The weights may then be adjusted to reduce the error. This manner of adjusting the weights may be referred to as “back propagation” as it involves a “backward pass” through the neural network”, i.e., ‘acquiring a distribution characteristic of the weight’)] [Lu: Paragraph 74 (“Quantize w.sub.l and h.sub.l Compute custom-character .sub.DJPQaccording to equation 20 Backward pass”, i.e., ‘determining, according to the distribution characteristic of the weight, an initial quantization parameter for quantizing the weight to reduce a quantization error …’)].
Ovtcharov et al, US 20200302269, [Ovtcharov: Abstract and paragraphs 9-10 and 50 (“The value of a loss function can then be computed at the end of each forward training pass of the ANN. The computed value for the loss function can then be used during a backward training pass (i.e. backpropagation) to compute a gradient for the first bit width (i.e. the bit width used to quantize weights) and to compute a gradient for the second bit width (i.e. the bit width used to quantize activation values)”, i.e., “bit width” is considered as ‘distribution characteristic of the weight’, which is then used to come up with “The value of the loss function” (‘an initial quantization parameter for quantizing the weight to reduce a quantization error’))] [Ovtcharov: Paragraph 33 (“In some examples, proprietary or open source libraries or frameworks are utilized to facilitate ANN creation, training 102, evaluation, and inference 104. Examples of such libraries include, but are not limited to, TENSORFLOW, MICROSOFT COGNITIVE TOOLKIT (“CNTK”), CAFFE, THEANO, and KERAS. In some examples, programming tools such as integrated development environments (“IDEs”) provide support for programmers and users to define, compile, and evaluate ANNs”)].
Chai et al, US 20200134461, [Chai: Paragraph 61 (“To better match the distribution of values between high-precision weights 114 and low-precision weights 116, machine learning system 104 may select, during training of DNN 106, a quantization function that best preserves the encoded distribution of the high-precision weights 114, even if the quantization is non-differentiable. For example, machine learning system 104 may use the following quantization function”, i.e., ‘distribution characteristic of the weight’)] [Chai: Paragraph 126 (“Alternatively, one can view BitNet as training with noisy gradients, an approach shown to encourage robust learning with few probability distributions for the noise. The noise incorporated in to the gradients of BitNet is similar to a disjoint set of gaussians with equally spaced centers determined by W and b, whose number is determined by b and variance by the range of values in W”, i.e., ‘adding noise to the quantized weight to obtain a noised weight’)] [Chai: Paragraph 19 (“Prior work known as BinaryConnect relates 1-bit precision parameters to a particular form of regularization. In particular, BinaryConnect is a method that trains a DNN with binary weights during forward and backward propagation, while retaining precision of the stored weights in which gradients are accumulated”)].
Suresh et al, US 20180089587, [Suresh: Paragraph 12 (“distributed mean estimation on data generated from a Gaussian distribution according to example embodiments”)].
Morrison et al, US 20230069536, [Morrison: Paragraphs 411 and 429 (“In some embodiments and/or usage scenarios, the weights are quantized, e.g., transformed to an integer data format. In some embodiments and/or usage scenarios, the integer data format is a reduced precision number format (e.g., 8-bit or 16-bit). The weights are then provided to one or more inference engines and used to make inferences in action”)].
Sriram et al, US 20220044114, [Sriram: Paragraphs 58, 63 and 109 (“QAT may be applied by quantizing all weights and activations of the network except for layers that require finer granularity in representation than the 8-bit quantization can provide (e.g., regression layers).” And “first applies QAT 108 by considering quantization errors when training a model. A training graph may be modified to simulate the lower precision behavior in the forward pass of the training process, and thus introduces the quantization errors as part of the training loss”, i.e., ‘to reduce a quantization error in quantizing the weight’)] [Sriram: Paragraph 113 (“In one or more embodiments, both weights and activations are uniformly quantized during the forward-pass of the training using the absolute maximum value of weights and a running average of absolute maximum value of activations over the training”, i.e., “absolute maximum value of weights and a running average of absolute maximum value of activations” is considered as ‘distribution characteristic of the weight’)] [Sriram: Paragraph 113 (“The gradients for weights and activations may be calculated via Straight-through estimation (STE) or some other estimate technique. These gradients are used to update the DNN weights during the backward-propagation pass”, i.e., “The gradients for weights and activations may be calculated via Straight-through estimation (STE)” is considered as ‘determining, according to the distribution characteristic of the weight, an initial quantization parameter for quantizing the weight to reduce an quantization error’)] [Sriram: Paragraph 487 (“a hub 3216, a crossbar (“XBar”) 3220, one or more general processing clusters”)].
Claim Rejections - 35 USC § 102
9. 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 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.
10. 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)(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.
11. Claims 1, 3, 5-6, 9-10, 12, 14-15, 18 and 20 are rejected under 35 U.S.C. 102(a)(2) as being anticipated by Sriram et al (US 20220044114).
Claim 1:
Sriram suggests a quantization method for a weight of a neural network [Sriram: Tilte (“Hybrid quantization of neural network”)], wherein the neural network is implemented on the basis of a crossbar-enabled analog computing-in-memory system [Sriram: Paragraph 487 (“a hub 3216, a crossbar (“XBar”) 3220, one or more general processing clusters”)], and the method comprises: acquiring a distribution characteristic of the weight [Sriram: Paragraph 113 (“In one or more embodiments, both weights and activations are uniformly quantized during the forward-pass of the training using the absolute maximum value of weights and a running average of absolute maximum value of activations over the training”, i.e., “absolute maximum value of weights and a running average of absolute maximum value of activations” is considered as ‘distribution characteristic of the weight’)]. Sriram suggests determining, according to the distribution characteristic of the weight, an initial quantization parameter for quantizing the weight to reduce a quantization error in quantizing the weight [Sriram: Paragraph 113 (“The gradients for weights and activations may be calculated via Straight-through estimation (STE) or some other estimate technique. These gradients are used to update the DNN weights during the backward-propagation pass”, i.e., “The gradients for weights and activations may be calculated via Straight-through estimation (STE)” is considered as ‘determining, according to the distribution characteristic of the weight, an initial quantization parameter for quantizing the weight to reduce an quantization error’)] [Sriram: Paragraphs 58, 63 and 109 (“QAT may be applied by quantizing all weights and activations of the network except for layers that require finer granularity in representation than the 8-bit quantization can provide (e.g., regression layers).” And “first applies QAT 108 by considering quantization errors when training a model. A training graph may be modified to simulate the lower precision behavior in the forward pass of the training process, and thus introduces the quantization errors as part of the training loss”, i.e., ‘to reduce a quantization error in quantizing the weight’)].
Claim 3:
Sriram suggests quantizing the weight using the initial quantization parameter to obtain a quantized weight [Sriram: Paragraph 113 (“In one or more embodiments, both weights and activations are uniformly quantized during the forward-pass of the training using the absolute maximum value of weights and a running average of absolute maximum value of activations over the training”, i.e., “absolute maximum value of weights and a running average of absolute maximum value of activations” is considered as ‘distribution characteristic of the weight’)]; and training the neural network using the quantized weight and updating the weight on the basis of a training result to obtain an updated weight [Sriram: Paragraphs 58, 63 and 109 “first applies QAT 108 by considering quantization errors when training a model. A training graph may be modified to simulate the lower precision behavior in the forward pass of the training process, and thus introduces the quantization errors as part of the training loss”)].
Claim 5:
Sriram suggests wherein training the neural network and updating the weight on the basis of the training result to obtain an updated weight comprise: performing forward propagation and backward propagation on the neural network; and updating the weight by using a gradient that is obtained by the backward propagation to obtain the updated weight [Sriram: Paragraph 113 (“In one or more embodiments, both weights and activations are uniformly quantized during the forward-pass of the training using the absolute maximum value of weights and a running average of absolute maximum value of activations over the training”, i.e., “absolute maximum value of weights and a running average of absolute maximum value of activations” is considered as ‘distribution characteristic of the weight’)] [Sriram: Paragraph 113 (“The gradients for weights and activations may be calculated via Straight-through estimation (STE) or some other estimate technique. These gradients are used to update the DNN weights during the backward-propagation pass”, i.e., “The gradients for weights and activations may be calculated via Straight-through estimation (STE)” is considered as ‘determining, according to the distribution characteristic of the weight, an initial quantization parameter for quantizing the weight to reduce an quantization error’)].
Claim 6:
Sriram suggests updating the initial quantization parameter on the basis of the updated weight [Sriram: Paragraph 113 (“The gradients for weights and activations may be calculated via Straight-through estimation (STE) or some other estimate technique. These gradients are used to update the DNN weights during the backward-propagation pass”, i.e., “The gradients for weights and activations may be calculated via Straight-through estimation (STE)” is considered as ‘determining, according to the distribution characteristic of the weight, an initial quantization parameter for quantizing the weight to reduce an quantization error’)].
Claim 9:
Claim 9 is essentially the same as claim 1 except that it sets forth the claimed invention as an apparatus rather than a method and rejected under the same reasons as applied above.
Claim 10:
Claim 10 is essentially the same as claim 3 except that it sets forth the claimed invention as an apparatus rather than a method and rejected under the same reasons as applied above.
Claim 12:
Claim 12 is essentially the same as claim 6 except that it sets forth the claimed invention as an apparatus rather than a method and rejected under the same reasons as applied above.
Claim 14:
Claim 14 is essentially the same as claim 1 except that it sets forth the claimed invention as an apparatus rather than a method and rejected under the same reasons as applied above.
Claim 15:
Claim 15 is essentially the same as claim 1 except that it sets forth the claimed invention as a program product rather than a method and rejected under the same reasons as applied above.
Claim 18:
Claim 18 is essentially the same as claim 5 and rejected under the same reasons as applied above.
Claim 20:
Claim 20 is essentially the same as claim 6 except that it sets forth the claimed invention as an apparatus rather than a method and rejected under the same reasons as applied above.
Claim Rejections - 35 USC § 103
12. 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 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.
13. 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.
14. Claims 2 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Sriram et al (US 20220044114), in view of Ovtcharov et al (US 20200302269).
Claim 2:
The combined teachings of Sriram and Ovtcharov suggest wherein determining, according to the distribution characteristic of the weight, the initial quantization parameter for quantizing the weight to reduce the quantization error in quantizing the weight comprises: acquiring a candidate distribution library, wherein multiple distribution models are stored
in the candidate distribution library; selecting, according to the distribution characteristic of the weight, a distribution model corresponding to the distribution characteristic from the candidate distribution library; and determining, according to the distribution model as selected, the initial quantization parameter for quantizing the weight to reduce the quantization error in quantizing the weight [Ovtcharov: Paragraph 33 (“In some examples, proprietary or open source libraries or frameworks are utilized to facilitate ANN creation, training 102, evaluation, and inference 104. Examples of such libraries include, but are not limited to, TENSORFLOW, MICROSOFT COGNITIVE TOOLKIT (“CNTK”), CAFFE, THEANO, and KERAS. In some examples, programming tools such as integrated development environments (“IDEs”) provide support for programmers and users to define, compile, and evaluate ANNs”)].
Both references (Sriram and Ovtcharov) taught features that were directed to analogous art and they were directed to the same field of endeavor, such as quantization of neural networks. It would have been obvious to one of ordinary skill in the art at the time the invention was made, having the teachings of Sriram and Ovtcharov before him/her, to modify the system of Sriram with the teaching of Ovtcharov in order to select proper libraries [Ovtcharov: Paragraph 33].
Claim 19:
Claim 19 is essentially the same as claim 2 except that it sets forth the claimed invention as an apparatus rather than a method and rejected under the same reasons as applied above.
15. Claims 4, 11 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Sriram et al (US 20220044114), in view of Chai et al (US 20200134461).
Claim 4:
The combined teachings of Sriram and Chai suggest quantizing the weight using the initial quantization parameter to obtain a quantized weight [Sriram: Paragraph 113 (“In one or more embodiments, both weights and activations are uniformly quantized during the forward-pass of the training using the absolute maximum value of weights and a running average of absolute maximum value of activations over the training”, i.e., “absolute maximum value of weights and a running average of absolute maximum value of activations” is considered as ‘distribution characteristic of the weight’)]; adding noise to the quantized weight to obtain a noised weight; and training the neural network using the noised weight and updating the weight on the basis of a training result to obtain an updated weight [Chai: Paragraph 126 (“Alternatively, one can view BitNet as training with noisy gradients, an approach shown to encourage robust learning with few probability distributions for the noise. The noise incorporated in to the gradients of BitNet is similar to a disjoint set of gaussians with equally spaced centers determined by W and b, whose number is determined by b and variance by the range of values in W”, i.e., ‘adding noise to the quantized weight to obtain a noised weight’)] [Chai: Paragraph 19 (“Prior work known as BinaryConnect relates 1-bit precision parameters to a particular form of regularization. In particular, BinaryConnect is a method that trains a DNN with binary weights during forward and backward propagation, while retaining precision of the stored weights in which gradients are accumulated”)].
Both references (Sriram and Chai) taught features that were directed to analogous art and they were directed to the same field of endeavor, such as quantization of neural networks. It would have been obvious to one of ordinary skill in the art at the time the invention was made, having the teachings of Sriram and Chai before him/her, to modify the system of Sriram with the teaching of Chai in order to add noise into consideration [Chai: Paragraph 126].
Claim 11:
Claim 11 is essentially the same as claim 4 except that it sets forth the claimed invention as an apparatus rather than a method and rejected under the same reasons as applied above.
Claim 17:
Claim 17 is essentially the same as claim 4 and rejected under the same reasons as applied above.
16. Claim 16 is rejected under 35 U.S.C. 103 as being unpatentable over Sriram et al (US 20220044114), in view of Ovtcharov et al (US 20200302269), and further in view of Chai et al (US 20200134461).
Claim 16:
The combined teachings of Sriram, Ovtcharov and Chai suggest quantizing the weight using the initial quantization parameter to obtain a quantized weight;
adding noise to the quantized weight to obtain a noised weight [Sriram: Paragraph 113 (“In one or more embodiments, both weights and activations are uniformly quantized during the forward-pass of the training using the absolute maximum value of weights and a running average of absolute maximum value of activations over the training”, i.e., “absolute maximum value of weights and a running average of absolute maximum value of activations” is considered as ‘distribution characteristic of the weight’)]; and training the neural network using the noised weight and updating the weight on the basis of a training result to obtain an updated weight [Chai: Paragraph 126 (“Alternatively, one can view BitNet as training with noisy gradients, an approach shown to encourage robust learning with few probability distributions for the noise. The noise incorporated in to the gradients of BitNet is similar to a disjoint set of gaussians with equally spaced centers determined by W and b, whose number is determined by b and variance by the range of values in W”, i.e., ‘adding noise to the quantized weight to obtain a noised weight’)] [Chai: Paragraph 19 (“Prior work known as BinaryConnect relates 1-bit precision parameters to a particular form of regularization. In particular, BinaryConnect is a method that trains a DNN with binary weights during forward and backward propagation, while retaining precision of the stored weights in which gradients are accumulated”)].
Three references (Sriram, Ovtcharov and Chai) taught features that were directed to analogous art and they were directed to the same field of endeavor, such as quantization of neural networks. It would have been obvious to one of ordinary skill in the art at the time the invention was made, having the teachings of Sriram, Ovtcharov and Chai before him/her, to modify the system of Sriram and Ovtcharov with the teaching of Chai in order to add noise into consideration [Chai: Paragraph 126].
Allowable Subject Matter
17. Claims 7-8 and 13 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
18. Any inquiry concerning this communication or earlier communications from the examiner should be directed to [Hung D. Le], whose telephone number is [571-270-1404]. The examiner can normally be communicated on [Monday to Friday: 9:00 A.M. to 5:00 P.M.].
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Apu Mofiz can be reached on [571-272-4080]. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, contact [800-786-9199 (IN USA OR CANADA) or 571-272-1000].
Hung Le
02/05/2026
/HUNG D LE/Primary Examiner, Art Unit 2161