DETAILED ACTION
Contents
Notice of Pre-AIA or AIA Status 2
Claim Rejections - 35 USC § 102 2
Claim Rejections - 35 USC § 103 5
Conclusion 16
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 .
This action is responsive to applicant’s claim set received on 6/18/24. Claims 1-16 are currently pending.
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 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, 2, 4 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Wang et al (CVPR: “Contrastive Learning based Hybrid Networks for Long-Tailed Image Classification”). Regarding claim 1, Wang discloses a method for training a neural network, the method comprising: obtaining first augmentation image features and second augmentation image features for a training image associated with a specified class (see section 2.2, 3.1, 3.2; contrastive learning using augmented views of the same image, labeling and normalized feature representations; Section 4.2, Implementation details: “Following SC loss, we also derive different views of an image by using different data augmentations in PSC loss. In our experiments, we simply use with and without color jitter as two different augmentation views.”; Section 3.3: “...force differently augmented views of each sample to be close to the prototype of their class...”; the specified class corresponds to the label y_i of the anchor image x_i, Section 3.3, Eq. (4): “z_i is the normalized representation of x_i”);
selecting, out of different sets of contrastive learning loss prototypes, a set of contrastive learning loss prototypes associated with the specified class (see 3.3; one prototype per class and extends to multiple prototypes per class);
wherein the different sets of contrastive learning loss prototypes are associated with different classes (see 3.3; m prototypes per class and identifies prototypes by class index; Section 3.3, “Extension to multiple prototypes per class,” Eq. (5): “where M is the number of prototypes per class, p_i^j denotes the representation for the i-th prototype of class j”; and Eq. (4): “p_{y_i} is the prototype representation for class y_i” – i.e., for a training image of class y_i, the set of M prototypes belonging to class y_i is selected from among the C class-specific sets of prototypes defined for classes j = 1,…,C; );
determining a contrastive learning loss on the first augmentation image features and the second augmentation image features, using the selected set of prototypes (see 3.3; PSC and MPSC loss functions using class protype or class-specific prototypes along with normalized feature representation; Section 3.3, Eq. (4): “L_PSC(z_i) = −log[exp(z_i·p_{y_i}/τ) / Σ_{j=1,j≠y_i}^{C} exp(z_i·p_j/τ)]”, computed on the differently-augmented-view features z_i using the selected class prototype p_{y_i}); and
updating the neural network based on the determined contrastive learning loss (see 3.1; final training combines supervised contrastive loss with cross-entropy loss; Section 4.2: “We use SGD with a momentum of 0.9 and weight decay of 1×10−4 as optimizer to train the hybrid networks”, wherein the hybrid loss of Eq. (1), which incorporates L_PSC/L_SCL, is minimized to update the parameters θ of the encoder network).
Regarding claim 2, Wang discloses determining one or more other losses, and wherein the updating of the neural network that generated the first network features and the second neural features is also based on the one or more other losses (Section 3.1: “The network consists of two branches: one contrastive learning branch for image representation learning and one cross-entropy driven branch for classifier learning”; Section 3.2, Eq. (1): “L_hybrid = α·L_SCL(B_SC) + (1−α)·L_CE(B_CE)”, wherein the cross-entropy classifier loss L_CE is determined and combined with the contrastive loss L_SCL, which includes the PSC loss of claim 1, to jointly update the network).
Regarding claim 4, Wang discloses determining of the constructive learning loss is responsive to values of the selected set of prototypes (Section 3.3, Eq. (4): “L_PSC(z_i) = −log[exp(z_i·p_{y_i}/τ)/Σ exp(z_i·p_j/τ)]”, wherein the computed loss value is a direct function of the dot-product similarity between the feature z_i and the values of the prototype vectors p_{y_i} and p_j).
Claim Rejections - 35 USC § 103
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.
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 claimedinvention is not identically disclosed as set forth in section 102 of this title, 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.
Claims 5, 8, 9, 10, 12, 13, 16 is rejected under 35 U.S.C. 103 as being unpatentable over Wang et al (CVPR: “Contrastive Learning based Hybrid Networks for Long-Tailed Image Classification”) in view of Coskun et al (US 2024/0087286 A1).
Regarding claim 5, Wang teaches all elements as mentioned above in claim 2. Wang does not teach expressly determining of the constructive learning loss is ignorant to values of the set of prototypes.
Coskun, in the same field of endeavor, teaches determining of the constructive learning loss is ignorant to values of the set of prototypes (paragraph [0115]: “The prototype clustering module 540 can apply a Sinkhorn-Knopp technique or algorithm to the features received from the first backbone network 530 to generate a first initial optimal assignment (OA) 550 that represents an initial classification of the features”; paragraph [0116]: similarly for the second set of features, “the prototype clustering module 540 can apply a Sinkhorn-Knopp technique or algorithm (or Cuturi formulation); paragraph [0119]: “the clustering system 500 can compute a first loss 570 based on a deviation between the OA generated based on the first set of features… and the updated first initial OA generated based on applying or using the second initial OA 552”; paragraph [0120]: “an additional loss term can be computed based on a deviation between the updated first initial OA… and the updated second initial OA).
It would have been obvious (before the effective filing date of the claimed invention) or (at the time the invention was made) to one of ordinary skill in the art to modify Wang to utilize the cited limitations as suggested by Coskun. The suggestion/motivation for doing so would have been to increase efficiency of classifying (see 0014). Furthermore, the prior art collectively includes each element claimed (though not all in the same reference), and one of ordinary skill in the art could have combined the elements in the manner explained above using known engineering design, interface and/or programming techniques, without changing a “fundamental” operating principle of Wang, while the teaching of Coskun continues to perform the same function as originally taught prior to being combined, in order to produce the repeatable and predictable result. It is for at least the aforementioned reasons that the examiner has reached a conclusion of obviousness with respect to the claim in question.
Regarding claim 8, Wang teaches all elements as mentioned above in claim 1. Wang does not teach expressly mapping the first neural network features to the set of prototypes to provide a first code, and mapping the second neural network features to the set of prototypes to provide a second code.
Coskun, in the same field of endeavor, teaches mapping the first neural network features to the set of prototypes to provide a first code, and mapping the second neural network features to the set of prototypes to provide a second code (Paragraph [0121] states: “the clustering system 500 receives minibatch of… videos X… and N prototypes P {p1,…, pN} represented by trainable vectors as the prototype clustering module 540. The clustering system 500 computes two random augmentations X’ and X^s… and computes their feature vectors F’={f1’,…,fN’} and F^s={f1^s,…,fN^s} using an encoder network θ… Next optimal assignments D’={d1’,…,dN’} and D^s={d1^s,…,dN^s} are computed by the prototype clustering module 540 from features F’ and F^s to prototypes P. di’∈D’ and di^s∈D^s vectors represent assignment values from fi’ and fi^s to prototypes, respectively.” That is, the first features F’ are mapped to the prototypes P to provide the first codes D’, and the second features F^s are mapped to the prototypes P to provide the second codes D^s).
It would have been obvious (before the effective filing date of the claimed invention) or (at the time the invention was made) to one of ordinary skill in the art to modify Wang to utilize the cited limitations as suggested by Coskun. The suggestion/motivation for doing so would have been to increase efficiency of classifying (see 0014). Furthermore, the prior art collectively includes each element claimed (though not all in the same reference), and one of ordinary skill in the art could have combined the elements in the manner explained above using known engineering design, interface and/or programming techniques, without changing a “fundamental” operating principle of Wang, while the teaching of Coskun continues to perform the same function as originally taught prior to being combined, in order to produce the repeatable and predictable result. It is for at least the aforementioned reasons that the examiner has reached a conclusion of obviousness with respect to the claim in question.
Regarding claim 9, Wang teaches a method for: obtaining first augmentation image features and second augmentation image features for a training image associated with a specified class (see section 2.2, 3.1, 3.2; contrastive learning using augmented views of the same image, labeling and normalized feature representations; Section 4.2, Implementation details: “Following SC loss, we also derive different views of an image by using different data augmentations in PSC loss. In our experiments, we simply use with and without color jitter as two different augmentation views.”; Section 3.3: “...force differently augmented views of each sample to be close to the prototype of their class...”; the specified class corresponds to the label y_i of the anchor image x_i, Section 3.3, Eq. (4): “z_i is the normalized representation of x_i”); selecting, out of different sets of contrastive learning loss prototypes, a set of contrastive learning loss prototypes associated with the specified class (see 3.3; one prototype per class and extends to multiple prototypes per class); wherein the different sets of contrastive learning loss prototypes are associated with different classes (see 3.3; m prototypes per class and identifies prototypes by class index; Section 3.3, “Extension to multiple prototypes per class,” Eq. (5): “where M is the number of prototypes per class, p_i^j denotes the representation for the i-th prototype of class j”; and Eq. (4): “p_{y_i} is the prototype representation for class y_i” – i.e., for a training image of class y_i, the set of M prototypes belonging to class y_i is selected from among the C class-specific sets of prototypes defined for classes j = 1,…,C; ); determining a contrastive learning loss on the first augmentation image features and the second augmentation image features, using the selected set of prototypes (see 3.3; PSC and MPSC loss functions using class protype or class-specific prototypes along with normalized feature representation; Section 3.3, Eq. (4): “L_PSC(z_i) = −log[exp(z_i·p_{y_i}/τ) / Σ_{j=1,j≠y_i}^{C} exp(z_i·p_j/τ)]”, computed on the differently-augmented-view features z_i using the selected class prototype p_{y_i}); and updating the neural network based on the determined contrastive learning loss (see 3.1; final training combines supervised contrastive loss with cross-entropy loss; Section 4.2: “We use SGD with a momentum of 0.9 and weight decay of 1×10−4 as optimizer to train the hybrid networks”, wherein the hybrid loss of Eq. (1), which incorporates L_PSC/L_SCL, is minimized to update the parameters θ of the encoder network). Wang does not teach expressly non-transitory computer readable medium for training a neural network, the non- transitory computer readable medium that stores instructions.
Coskun, in the same field of endeavor, teaches non-transitory computer readable medium for training a neural network, the non- transitory computer readable medium that stores instructions (see 0129, 0133-0135, 0144; non-transitory CRM).
It would have been obvious (before the effective filing date of the claimed invention) or (at the time the invention was made) to one of ordinary skill in the art to modify Wang to utilize the cited limitations as suggested by Coskun. The suggestion/motivation for doing so would have been to increase efficiency of classifying (see 0014). Furthermore, the prior art collectively includes each element claimed (though not all in the same reference), and one of ordinary skill in the art could have combined the elements in the manner explained above using known engineering design, interface and/or programming techniques, without changing a “fundamental” operating principle of Wang, while the teaching of Coskun continues to perform the same function as originally taught prior to being combined, in order to produce the repeatable and predictable result. It is for at least the aforementioned reasons that the examiner has reached a conclusion of obviousness with respect to the claim in question.
Regarding claims 13, 16, the claims are analyzed as a non-transitory CRM that implements the limitations of claims 5 and 8 (see rejection of claims 5 and 8).
Regarding claim 10, Wang discloses determining one or more other losses, and wherein the updating of the neural network that generated the first network features and the second neural features is also based on the one or more other losses (Section 3.1: “The network consists of two branches: one contrastive learning branch for image representation learning and one cross-entropy driven branch for classifier learning”; Section 3.2, Eq. (1): “L_hybrid = α·L_SCL(B_SC) + (1−α)·L_CE(B_CE)”, wherein the cross-entropy classifier loss L_CE is determined and combined with the contrastive loss L_SCL, which includes the PSC loss of claim 1, to jointly update the network).
Regarding claim 12, Wang discloses determining of the constructive learning loss is responsive to values of the selected set of prototypes (Section 3.3, Eq. (4): “L_PSC(z_i) = −log[exp(z_i·p_{y_i}/τ)/Σ exp(z_i·p_j/τ)]”, wherein the computed loss value is a direct function of the dot-product similarity between the feature z_i and the values of the prototype vectors p_{y_i} and p_j).
Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over Wang et al (CVPR: “Contrastive Learning based Hybrid Networks for Long-Tailed Image Classification”) in view of Zhu et al (MDPI: “Supervised Contrastive Learning with Angular Margin for the Detection and Grading of Diabetic Retinopathy”).
Regarding claim 3, Wang teaches all elements as mentioned above in claim 2. Wang does not teach expressly one or more angular related losses for the first augmented image and for the second augmented image.
Zhu, in the same field of endeavor, teaches one or more angular related losses for the first augmented image and for the second augmented image (Section 3.3: “The collection of N data is referred to as a ‘batch’, and the set of 2N augmented data is referred to as a ‘double-viewed batch’.” Zhu, Eq. (3): “θ_{i,p} = arccos(z_i·z_p / (‖z_i‖∗‖z_p‖))”, converts the feature similarity into an angular representation. Zhu, Eq. (4): “L^{a-m}_i = −Σ_{p∈P(i)} (1/|P(i)|) log[exp(cos(θ_{i,p}+I_u^p m_u+I_v^p m_v)/τ) / Σ_{a∈A(i)} exp(cos(θ_{i,a}+U·I_u^a m_u+V·I_v^a m_v)/τ)]”, is applied over both the augmented-positive (U, corresponding to the first/second augmented images of the same training sample) and same-label-positive (V) pairs of the double-viewed batch. Zhu, Eq. (6): “L = L_ce + λL^{a-m}).
It would have been obvious (before the effective filing date of the claimed invention) or (at the time the invention was made) to one of ordinary skill in the art to modify Wang to utilize the cited limitations as suggested by Zhu. The suggestion/motivation for doing so would have been to better the classifiers (see abstract). Furthermore, the prior art collectively includes each element claimed (though not all in the same reference), and one of ordinary skill in the art could have combined the elements in the manner explained above using known engineering design, interface and/or programming techniques, without changing a “fundamental” operating principle of Wang, while the teaching of Zhu continues to perform the same function as originally taught prior to being combined, in order to produce the repeatable and predictable result. It is for at least the aforementioned reasons that the examiner has reached a conclusion of obviousness with respect to the claim in question.
Claim 11 is rejected under 35 U.S.C. 103 as being unpatentable over Wang et al (CVPR: “Contrastive Learning based Hybrid Networks for Long-Tailed Image Classification”) with Coskun et al (US 2024/0087286 A1), and further in view of Zhu et al (MDPI: “Supervised Contrastive Learning with Angular Margin for the Detection and Grading of Diabetic Retinopathy”).
Regarding claim 11, Wang with Coskun teaches all elements as mentioned above in claim 10. Wang with Coskun does not teach expressly one or more angular related losses for the first augmented image and for the second augmented image.
Zhu, in the same field of endeavor, teaches one or more angular related losses for the first augmented image and for the second augmented image (Section 3.3: “The collection of N data is referred to as a ‘batch’, and the set of 2N augmented data is referred to as a ‘double-viewed batch’.” Zhu, Eq. (3): “θ_{i,p} = arccos(z_i·z_p / (‖z_i‖∗‖z_p‖))”, converts the feature similarity into an angular representation. Zhu, Eq. (4): “L^{a-m}_i = −Σ_{p∈P(i)} (1/|P(i)|) log[exp(cos(θ_{i,p}+I_u^p m_u+I_v^p m_v)/τ) / Σ_{a∈A(i)} exp(cos(θ_{i,a}+U·I_u^a m_u+V·I_v^a m_v)/τ)]”, is applied over both the augmented-positive (U, corresponding to the first/second augmented images of the same training sample) and same-label-positive (V) pairs of the double-viewed batch. Zhu, Eq. (6): “L = L_ce + λL^{a-m}).
It would have been obvious (before the effective filing date of the claimed invention) or (at the time the invention was made) to one of ordinary skill in the art to modify Wang with Coskun to utilize the cited limitations as suggested by Zhu. The suggestion/motivation for doing so would have been to better the classifiers (see abstract). Furthermore, the prior art collectively includes each element claimed (though not all in the same reference), and one of ordinary skill in the art could have combined the elements in the manner explained above using known engineering design, interface and/or programming techniques, without changing a “fundamental” operating principle of Wang with Coskun, while the teaching of Zhu continues to perform the same function as originally taught prior to being combined, in order to produce the repeatable and predictable result. It is for at least the aforementioned reasons that the examiner has reached a conclusion of obviousness with respect to the claim in question.
Claims 6-7 are rejected under 35 U.S.C. 103 as being unpatentable over Wang et al (CVPR: “Contrastive Learning based Hybrid Networks for Long-Tailed Image Classification”) in view of Caron et al (NIPS: “Unsupervised Learning of Visual Features by Contrasting Cluster Assignments”).
Regarding claims 6-7, Wang teaches all elements as mentioned above in claim 1. Wang does not teach expressly (i) determining a first code associated with the first augmented image, (ii) determining a second code associated with the second augmented image, (iii) determining a first code estimate based on the second neural network features, (iv) determining a second code estimate based on the first neural network features;
(v) determining a first fit metric based on a fit between the first code and the first code estimate, (vi) determining a second fit metric based on a fit between the second code and the second code estimate, and (v) determining the constructive learning loss based on the first fit metric and the second fit metric.
Caron, in the same field of endeavor, teaches (i) determining a first code associated with the first augmented image, (ii) determining a second code associated with the second augmented image, (iii) determining a first code estimate based on the second neural network features, (iv) determining a second code estimate based on the first neural network features (section 3: Given two image features z_t and z_s from two different augmentations of the same image, we compute their codes q_t and q_s by matching these features to a set of K prototypes {c1,…,cK}. We then setup a ‘swapped’ prediction problem with the following loss function: L(z_t, z_s) = ℓ(z_t, q_s) + ℓ(z_s, q_t)” (Eq. 1)). That is, a first code q_t is determined for the first augmented image (feature z_t), and a second code q_s is determined for the second augmented image (feature z_s). Caron, Eq. (2): “ℓ(z_t,q_s) = −Σ_k q_s^{(k)} log p_t^{(k)}”, where p_t^{(k)} = exp((1/τ)z_tᵀc_k)/Σ_{k’} exp((1/τ)z_tᵀc_{k’}) is an estimate, based on the first neural network features z_t, of the second code q_s (i.e., a second code estimate based on the first neural network features); symmetrically, the term ℓ(z_s,q_t) uses p_s, an estimate based on the second neural network features z_s, of the first code q_t (i.e., a first code estimate based on the second neural network features);
(v) determining a first fit metric based on a fit between the first code and the first code estimate, (vi) determining a second fit metric based on a fit between the second code and the second code estimate, and (v) determining the constructive learning loss based on the first fit metric and the second fit metric (section 3: where the function ℓ(z,q) measures the fit between features z and a code q, as detailed later”; Eq. (1): “L(z_t,z_s) = ℓ(z_t,q_s) + ℓ(z_s,q_t)”, wherein ℓ(z_s,q_t) is a first fit metric based on the fit between the first code q_t and the first code estimate p_s (computed from z_s), ℓ(z_t,q_s) is a second fit metric based on the fit between the second code q_s and the second code estimate p_t (computed from z_t), and the overall contrastive loss L(z_t,z_s) is determined as the sum of the first and second fit metrics).
It would have been obvious (before the effective filing date of the claimed invention) or (at the time the invention was made) to one of ordinary skill in the art to modify Wang to utilize the cited limitations as suggested by Caron. The suggestion/motivation for doing so would have been to enhance the accuracy (see abstract). Furthermore, the prior art collectively includes each element claimed (though not all in the same reference), and one of ordinary skill in the art could have combined the elements in the manner explained above using known engineering design, interface and/or programming techniques, without changing a “fundamental” operating principle of Wang, while the teaching of Caron continues to perform the same function as originally taught prior to being combined, in order to produce the repeatable and predictable result. It is for at least the aforementioned reasons that the examiner has reached a conclusion of obviousness with respect to the claim in question.
Claims 14-15 are rejected under 35 U.S.C. 103 as being unpatentable over Wang et al (CVPR: “Contrastive Learning based Hybrid Networks for Long-Tailed Image Classification”) with Coskun et al (US 2024/0087286 A1), and further in view of Caron et al (NIPS: “Unsupervised Learning of Visual Features by Contrasting Cluster Assignments”).
Regarding claims 14-15, Wang with Coskun teaches all elements as mentioned above in claim 9. Wang with Coskun does not teach expressly (i) determining a first code associated with the first augmented image, (ii) determining a second code associated with the second augmented image, (iii) determining a first code estimate based on the second neural network features, (iv) determining a second code estimate based on the first neural network features;
(v) determining a first fit metric based on a fit between the first code and the first code estimate, (vi) determining a second fit metric based on a fit between the second code and the second code estimate, and (v) determining the constructive learning loss based on the first fit metric and the second fit metric.
Caron, in the same field of endeavor, teaches (i) determining a first code associated with the first augmented image, (ii) determining a second code associated with the second augmented image, (iii) determining a first code estimate based on the second neural network features, (iv) determining a second code estimate based on the first neural network features (section 3: Given two image features z_t and z_s from two different augmentations of the same image, we compute their codes q_t and q_s by matching these features to a set of K prototypes {c1,…,cK}. We then setup a ‘swapped’ prediction problem with the following loss function: L(z_t, z_s) = ℓ(z_t, q_s) + ℓ(z_s, q_t)” (Eq. 1)). That is, a first code q_t is determined for the first augmented image (feature z_t), and a second code q_s is determined for the second augmented image (feature z_s). Caron, Eq. (2): “ℓ(z_t,q_s) = −Σ_k q_s^{(k)} log p_t^{(k)}”, where p_t^{(k)} = exp((1/τ)z_tᵀc_k)/Σ_{k’} exp((1/τ)z_tᵀc_{k’}) is an estimate, based on the first neural network features z_t, of the second code q_s (i.e., a second code estimate based on the first neural network features); symmetrically, the term ℓ(z_s,q_t) uses p_s, an estimate based on the second neural network features z_s, of the first code q_t (i.e., a first code estimate based on the second neural network features);
(v) determining a first fit metric based on a fit between the first code and the first code estimate, (vi) determining a second fit metric based on a fit between the second code and the second code estimate, and (v) determining the constructive learning loss based on the first fit metric and the second fit metric (section 3: where the function ℓ(z,q) measures the fit between features z and a code q, as detailed later”; Eq. (1): “L(z_t,z_s) = ℓ(z_t,q_s) + ℓ(z_s,q_t)”, wherein ℓ(z_s,q_t) is a first fit metric based on the fit between the first code q_t and the first code estimate p_s (computed from z_s), ℓ(z_t,q_s) is a second fit metric based on the fit between the second code q_s and the second code estimate p_t (computed from z_t), and the overall contrastive loss L(z_t,z_s) is determined as the sum of the first and second fit metrics).
It would have been obvious (before the effective filing date of the claimed invention) or (at the time the invention was made) to one of ordinary skill in the art to modify Wang with Coskun to utilize the cited limitations as suggested by Caron. The suggestion/motivation for doing so would have been to enhance the accuracy (see abstract). Furthermore, the prior art collectively includes each element claimed (though not all in the same reference), and one of ordinary skill in the art could have combined the elements in the manner explained above using known engineering design, interface and/or programming techniques, without changing a “fundamental” operating principle of Wang with Coskun, while the teaching of Caron continues to perform the same function as originally taught prior to being combined, in order to produce the repeatable and predictable result. It is for at least the aforementioned reasons that the examiner has reached a conclusion of obviousness with respect to the claim in question.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to EDWARD PARK. The examiner’s contact information is as follows:
Telephone: (571)270-1576 | Fax: 571.270.2576 | Edward.Park@uspto.gov
For email communications, please notate MPEP 502.03, which outlines procedures pertaining to communications via the internet and authorization. A sample authorization form is cited within MPEP 502.03, section II.
The examiner can normally be reached on M-F 9-6 CST.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, John M. Villecco, can be reached on (571) 272-7319. 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, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/EDWARD PARK/ Primary Examiner, Art Unit 2675
Prosecution Insights