DETAILED ACTION
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 . In communications filed on 07/08/2015. Claims 1-8 are pending in this examination.
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. This examination is in response to US Patent Application No. 19/262,095.
Double Patenting
The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory obviousness-type double patenting rejection is appropriate where the conflicting claims are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); and In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on a nonstatutory double patenting ground provided the conflicting application or patent either is shown to be commonly owned with this application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement.
Claims 1-8 are rejected under the judicially created doctrine of obviousness-type double patenting as being unpatentable over claims 1-5, respectively, of U.S. Patent No. 12387002 , and Claims 1-6 , respectfully, of U.S. Co-pending Application No. 19/262,090.
Although the conflicting claims are not identical, they are not patentably distinct from each other because:
The patent claims include all of the limitations of the instant application claims, respectively. The patent claims also include additional limitations. Hence, the instant application claims are generic to the species of invention covered by the respective patent claims. As such, the instant application claims are anticipated by the patent claims and are therefore not patentably distinct therefrom. (See Eli Lilly and Co. v. Barr Laboratories Inc., 58 USPQ2D 1869, "a later genus claim limitation is anticipated by, and therefore not patentably distinct from, an earlier species claim", In re Goodman, 29 USPQ2d 2010, "Thus, the generic invention is 'anticipated' by the species of the patented invention" and the instant “application claims are generic to species of invention covered by the patent claim, and since without terminal disclaimer, extant species claims preclude issuance of generic application claims”).
Patented Application 12387002
Instant Application 16817831
Co-pending Application 19262090
A system comprising: a secure in-memory unit implemented on an associative processing unit (APU), said APU implemented in a memory array having columns, for creating encrypted vectors for secure data transfer to an external similarity searcher, said unit to implement: an in-memory secure data store to store raw data to be kept secure in said columns; an in-memory neural proxy hash encoder comprising a trained neural network trained to encode data into binary encoded feature sets having a non-recoverable representation of said raw data, said encoder operative in user-selected ones of said columns to encode user selected ones of said raw data into a binary encoded feature set stored in portions of said columns; and an in-memory encryptor operative in said portions of said columns to encrypt said binary encoded feature set into an encrypted vector for said secure data transfer.
The system of claim 1 said trained neural network to encode at least one of: image files, audio files, and large data sets.
The system of claim 1 wherein said APU is implemented on one of: SRAM, non-volatile, and non-destructive memory.
A system comprising: a secure, in-memory unit implemented on an associative processing unit (APU), said APU implemented in a memory array having columns, for securely receiving and decrypting data to be kept secure during a similarity search, said unit to implement: an in-memory decryptor to decrypt a received encrypted data vector into a secure encoded data vector formed of a binary encoded feature set generated by an external neural proxy hash encoder, wherein said encoded feature set is a non-recoverable representation of user selected raw data to be kept secure; and an in-memory encoded vector data store to store said secure encoded data vector in said columns among a plurality of previously decrypted secure encoded data vectors, said secure encoded vector to act as a query vector to said similarity search through said previously decrypted secure encoded data vectors.
The system of claim 4 wherein said APU is implemented on one of: SRAM, non-volatile, and non-destructive memory.
1. A system comprising: a secure, in-memory unit implemented on an associative processing unit ( APU), for performing a secure similarity search, said unit to implement: a decryptor to decrypt an encrypted data vector into a data vector; a neural proxy hash encoder to encode said data vector into an encoded search data vector; an encoded vector data store to store a plurality of encoded search candidate vectors; and a similarity searcher to perform a similarity search between said encoded search query vector and said plurality of encoded search candidate vectors.
2. The system of claim I wherein said encoded search data vector is one of: an encoded search query vector and an encoded search candidate vector.
3. The system of claim 1 said vector data store to store said encoded search candidate vectors in columns.
4. The system of claim 3 said similarity searcher to perform said similarity search of said plurality of encoded search candidate vectors in said columns in a parallel process.
5. The system of claim 1 wherein said similarity search is a nearest neighbor search.
6. The system of claim 1 wherein said neural proxy hash encoder comprises a trained neural network comprising a plurality of layers to encode input data into feature sets.
7. The system of claim 6 said trained neural network to encode at least one of: image files, audio files, and large data set files.
8. The system of claim I wherein said APU is implemented on one of: SRAM, non-volatile, and non-destructive memory.
1. A system comprising: a secure, in-memory unit implemented on an associative processing unit (APU), for performing a secure similarity search, said unit to implement: a decryptor to decrypt an encrypted, encoded vector into an encoded vector; an encoded vector data store to store a plurality of encoded search candidate vectors; and a similarity searcher to perform a similarity search between an encoded search query vector and said plurality of encoded search candidate vectors.
2. The system of claim 1 wherein said encoded vector is one of: an encoded search query vector and an encoded search candidate vector.
3. The system of claim 1wherein said vector data store to store said encoded search candidate vectors in columns.
4. The system of claim 3 said similarity searcher to perform said similarity search of said plurality of encoded search candidate vectors in said columns in a parallel process.
5. The system of claim 1wherein said similarity search is a nearest neighbor search.
6. The system of claim 1wherein said APU implemented on one of: SRAM, non-volatile, and non-destructive memory.
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 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.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied 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.
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 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 1-5, and 7-8 are rejected under 35 U.S.C. 103 as being unpatentable over US Patent Application No. (US US20180341642) (filed in IDS 11/06/2025) issued to Akerib and in view of US Patent No. Application (US2021/0342700) issued to ANTIC(filed in IDS 11/06/2025), and furthermore in view of DAUB (US2021/0004864) (filed in IDS 11/06/2025).
Regarding claim 1, Akerib discloses a secure, in memory unit implemented on an associative unit ( APU), for performing a secure similarity search, said unit to implement:
[10, there is therefore provided, in accordance with a preferred embodiment of the present invention, a system for natural language processing. The system includes a memory array and an in-memory processor. The memory array has rows and columns and is divided into a similarity section initially storing a plurality of feature or key vectors, a SoftMax section in which to determine probabilities of occurrence of the feature or key vectors, a value section initially storing a plurality of modified feature vectors, and a marker section. Operations in one or more columns of the memory array are associated with one feature vector to be processed], and [¶51, Reference is now made to FIGS. 1A and 1B, which are schematic illustrations of a memory computation device 100, constructed and operative in accordance with a preferred embodiment of the present invention. As illustrated in FIG. 1A, device 100 may comprise a memory array 110 to store a dataset, a k-Mins processor 120, implemented on a memory logic element, to perform a k-Mins operation and a k-Mins temporary store 130 that may be used for storing intermediate and final results of operations made by k-Mins processor 120 on data stored in memory array 110. In FIG. 1B the physical aspects of k-Mins processor 120 and the k-Mins temporary store 130 are illustrated in associative memory array 140. Associative memory array 140 combines the operations of k-Mins processor 120 and the store of k-Mins temporary store 130. Memory array 110 may store a very large dataset of binary numbers. Each binary number is comprised of a fixed number of bits and is stored in a different column in memory array 110. K-Mins temporary store 120 may store copies of the information stored in memory array 110 and several vectors storing temporary information related to a step of the computation performed by k-Mins processor 120 as well as the final result including an indication of k columns storing the k lowest values in the dataset], and [ ¶115, Applicant has realized that an associative processing unit (APU), such as memory computation device 100, can do any function of search, such as cosine similarity which is not an exact match, to achieve all that is needed for natural language processing with a neural network]; and
an encoded vector data store to store a plurality of encoded search candidate vectors [ Abstract, a system for natural language processing includes a memory array and a processor. The memory array is divided into a similarity section storing a plurality of feature vectors, a SoftMax section in which to determine probabilities of occurrence of the feature vectors, a value section storing a plurality of modified feature vectors, and a marker section. The processor activates the array to perform parallel operations in each column indicated by the marker section: a similarity operation in the similarity section between a vector question and feature vectors stored in indicated columns; a SoftMax operation in the SoftMax section to determine an associated SoftMax probability value for indicated feature vectors; a multiplication operation in the value section to multiply the associated SoftMax value by modified feature vectors stored in indicated columns; and a vector sum in the value section to accumulate an attention vector of output of the multiplication operation]; and
and a similarity searcher to perform a similarity search between said encoded search query vector and said plurality of encoded search candidate vectors [¶10, There is therefore provided, in accordance with a preferred embodiment of the present invention, a system for natural language processing. The system includes a memory array and an in-memory processor. The memory array has rows and columns and is divided into a similarity section initially storing a plurality of feature or key vectors, a SoftMax section in which to determine probabilities of occurrence of the feature or key vectors, a value section initially storing a plurality of modified feature vectors, and a marker section. Operations in one or more columns of the memory array are associated with one feature vector to be processed], and [¶51, Reference is now made to FIGS. 1A and 1B, which are schematic illustrations of a memory computation device 100, constructed and operative in accordance with a preferred embodiment of the present invention. As illustrated in FIG. 1A, device 100 may comprise a memory array 110 to store a dataset, a k-Mins processor 120, implemented on a memory logic element, to perform a k-Mins operation and a k-Mins temporary store 130 that may be used for storing intermediate and final results of operations made by k-Mins processor 120 on data stored in memory array 110. In FIG. 1B the physical aspects of k-Mins processor 120 and the k-Mins temporary store 130 are illustrated in associative memory array 140. Associative memory array 140 combines the operations of k-Mins processor 120 and the store of k-Mins temporary store 130. Memory array 110 may store a very large dataset of binary numbers. Each binary number is comprised of a fixed number of bits and is stored in a different column in memory array 110. K-Mins temporary store 120 may store copies of the information stored in memory array 110 and several vectors storing temporary information related to a step of the computation performed by k-Mins processor 120 as well as the final result including an indication of k columns storing the k lowest values in the dataset], and [ ¶115, Applicant has realized that an associative processing unit (APU), such as memory computation device 100, can do any function of search, such as cosine similarity which is not an exact match, to achieve all that is needed for natural language processing with a neural network].
Akerib does not explicitly disclose, however, ANTIC discloses a decryptor to decrypt an encrypted data vector into a data vector [0031, Referring now to FIG. 3, a functional block diagram of a system 300 (analogous to system 200) for training an AI model 304 for performing deterministic data processing is illustrated, in accordance with an embodiment. It will be apparent to a person skilled in the art that the system 200 and the system 300 may be combined to form a single system. The system 300 may include a data processing device 302 (analogous to the data processing device 202) and the AI model 304 (analogous to the AI model 204). In order to train the AI model 304, the data processing device 302 may include a deterministic data processing model 306, a feature vector generating module 308, and a feedback module 312. The feature vector generating module 308 may further include a deep learning network 310], and [¶¶38-39, Referring back to FIG. 3, in some embodiments, the feature vector generating module 308 may further include an encryption module 316. The encryption module 316 may encrypt the training set of input feature vectors and the training set of output feature vectors based on Homomorphic keys. It may be noted that in such embodiments, the AI model 304 may be trained using encrypted training set of input vectors and encrypted training set of output vectors… Referring back to FIG. 3, in some embodiments, the feature vector generating module 308 may further include an encryption module 316. The encryption module 316 may encrypt the training set of input feature vectors and the training set of output feature vectors based on Homomorphic keys. It may be noted that in such embodiments, the AI model 304 may be trained using encrypted training set of input vectors and encrypted training set of output vectors].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the teaching of Akerib with the teaching of ANTIC in order to performing deterministic data processing through Artificial Intelligence (AI). The techniques provide for homomorphic encrypting of feature vector data extracted from input and output data, which makes the feature vector data safer to be transmitted and stored in memory or disk [ ANTIC, ¶56].
Akerib does not explicitly disclose, however, while ANTIC discloses a neural proxy hash encoder to encode said data vector into an encoded search data vector
[ Abstract, A method for performing deterministic data processing through Artificial Intelligence (AI) is disclosed. The method may include generating, via a deep learning network, a set of input feature vectors based on input data for a deterministic data processing model. The method may further include providing the set of input feature vectors to a trained AI model. The trained AI model may generate a set of output feature vectors that may correspond to an output data of the deterministic data processing model. The method may further include determining a variation between the set of output feature vectors and the output data, and iteratively performing incremental learning of the AI model based on the determined variation], and [0031, Referring now to FIG. 3, a functional block diagram of a system 300 (analogous to system 200) for training an AI model 304 for performing deterministic data processing is illustrated, in accordance with an embodiment. It will be apparent to a person skilled in the art that the system 200 and the system 300 may be combined to form a single system. The system 300 may include a data processing device 302 (analogous to the data processing device 202) and the AI model 304 (analogous to the AI model 204). In order to train the AI model 304, the data processing device 302 may include a deterministic data processing model 306, a feature vector generating module 308, and a feedback module 312. The feature vector generating module 308 may further include a deep learning network 310], and [¶¶38-39, Referring back to FIG. 3, in some embodiments, the feature vector generating module 308 may further include an encryption module 316. The encryption module 316 may encrypt the training set of input feature vectors and the training set of output feature vectors based on Homomorphic keys. It may be noted that in such embodiments, the AI model 304 may be trained using encrypted training set of input vectors and encrypted training set of output vectors… Referring back to FIG. 3, in some embodiments, the feature vector generating module 308 may further include an encryption module 316. The encryption module 316 may encrypt the training set of input feature vectors and the training set of output feature vectors based on Homomorphic keys. It may be noted that in such embodiments, the AI model 304 may be trained using encrypted training set of input vectors and encrypted training set of output vectors], and [ ¶¶55-51, … a dense Artificial Neural Network (ANN) part in the middle may also be multi-layered…].
Furthermore, DAUB discloses [¶8, At least another aspect is directed to a method for providing anonymous data about user interactions with a set of content items. The method includes receiving, via a network, data to identify a set of hash function. The method further includes retrieving a plurality of user identifiers, each of the user identifiers identifying interaction with a set of content items by a respective user of the plurality. The method further includes for each of the plurality of user identifiers, generating k hashes of the user identifier, each hash corresponding to one of the sets of hash functions, and generating a user vector, each coordinate of the user vector corresponding to a bit value of a respective hash. The method also includes generating an interaction vector by summing the plurality of generated user vectors. The method further includes transmitting, via a network, the generated interaction vector via a network to the requesting party], and [¶84, In yet another approach, the user identifiers can be encrypted or hashed prior to generating the vectors discussed above in relation to FIGS. 3-4B. For example, the second data processing system 104 and the third data processing system 106 can encrypt or compute a hash of each of the user identities in the first set of user identities 302 and the second set of user identities 304. The second data processing system 104 and the third data processing system 106 can then generate the vectors (the binomial vectors or the vectors of counts) based on the encrypted set of user identities. Encrypting the user identifiers before generating the vectors can improve the privacy of the user identities. In some instances, where vectors are formed from the same user identities, the intersection of the vectors may still include some information related to the user identities. By encrypting or hashing the user identities prior to generating the vectors, the risk of leakage of information can be reduced.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the teaching of Akerib, and ANTIC with the teaching of DAUB in order to improve the privacy of the user identities by encrypting or hashing the user identities prior to generating the vectors, the risk of leakage of information can be reduced [ DAUB, ¶84].
Regarding claim 2, the system of claim 1 wherein said encoded search data vector is one of: an encoded search query vector and an encoded search candidate vector [ Abstract, a system for natural language processing includes a memory array and a processor. The memory array is divided into a similarity section storing a plurality of feature vectors, a SoftMax section in which to determine probabilities of occurrence of the feature vectors, a value section storing a plurality of modified feature vectors, and a marker section. The processor activates the array to perform parallel operations in each column indicated by the marker section: a similarity operation in the similarity section between a vector question and feature vectors stored in indicated columns; a SoftMax operation in the SoftMax section to determine an associated SoftMax probability value for indicated feature vectors; a multiplication operation in the value section to multiply the associated SoftMax value by modified feature vectors stored in indicated columns; and a vector sum in the value section to accumulate an attention vector of output of the multiplication operation].
Regarding claim 3, Akerib discloses the system of claim 1 said vector data store to store said encoded search candidate vectors in columns [ Abstract, a system for natural language processing includes a memory array and a processor. The memory array is divided into a similarity section storing a plurality of feature vectors, a SoftMax section in which to determine probabilities of occurrence of the feature vectors, a value section storing a plurality of modified feature vectors, and a marker section. The processor activates the array to perform parallel operations in each column indicated by the marker section: a similarity operation in the similarity section between a vector question and feature vectors stored in indicated columns; a SoftMax operation in the SoftMax section to determine an associated SoftMax probability value for indicated feature vectors; a multiplication operation in the value section to multiply the associated SoftMax value by modified feature vectors stored in indicated columns; and a vector sum in the value section to accumulate an attention vector of output of the multiplication operation.
[0010] There is therefore provided, in accordance with a preferred embodiment of the present invention, a system for natural language processing. The system includes a memory array and an in-memory processor. The memory array has rows and columns and is divided into a similarity section initially storing a plurality of feature or key vectors, a SoftMax section in which to determine probabilities of occurrence of the feature or key vectors, a value section initially storing a plurality of modified feature vectors, and a marker section. Operations in one or more columns of the memory array are associated with one feature vector to be processed. The in-memory processor activates the memory array to perform the following operations in parallel in each column indicated by the marker section.
Regarding claim 4 , Akerib discloses the system of claim 3 said similarity searcher to perform said similarity search of said plurality of encoded search candidate vectors in said columns in a parallel process[ Abstract, a system for natural language processing includes a memory array and a processor. The memory array is divided into a similarity section storing a plurality of feature vectors, a SoftMax section in which to determine probabilities of occurrence of the feature vectors, a value section storing a plurality of modified feature vectors, and a marker section. The processor activates the array to perform parallel operations in each column indicated by the marker section: a similarity operation in the similarity section between a vector question and feature vectors stored in indicated columns; a SoftMax operation in the SoftMax section to determine an associated SoftMax probability value for indicated feature vectors; a multiplication operation in the value section to multiply the associated SoftMax value by modified feature vectors stored in indicated columns; and a vector sum in the value section to accumulate an attention vector of output of the multiplication operation.
[0010] There is therefore provided, in accordance with a preferred embodiment of the present invention, a system for natural language processing. The system includes a memory array and an in-memory processor. The memory array has rows and columns and is divided into a similarity section initially storing a plurality of feature or key vectors, a SoftMax section in which to determine probabilities of occurrence of the feature or key vectors, a value section initially storing a plurality of modified feature vectors, and a marker section. Operations in one or more columns of the memory array are associated with one feature vector to be processed. The in-memory processor activates the memory array to perform the following operations in parallel in each column indicated by the marker section.
Regarding claim 5 , Akerib discloses the system of claim 1 wherein said similarity search is a nearest neighbor search [0097] The k-Mins algorithm described hereinabove may be used by the k nearest neighbors (K-NN) data mining algorithm. In K-NN D may represent a large dataset containing q objects (q enormously large). D.sup.P is one object in dataset D: D.sup.P E D and A is the object to classify. An object is defined by a vector of numerical attributes: A is defined by a vector [A.sub.0, A.sub.1, . . . A.sub.n] of n attributes and D.sup.P is defined by a vector [D.sub.1.sup.p, D.sub.2.sup.p, D.sub.3.sup.p, . . . D.sub.n.sup.p] of the same n attributes. A distance, which is a binary number C.sup.P of m bits, between object A and object D.sup.P, is calculated between the introduced object A and each object D.sup.P in the dataset D. The distance C.sup.P may represent the cosine similarity between two non-zero vectors. The known in the art cosine similarity associate
Regarding claim 7, Akerib does not explicitly disclose, however, ANTC discloses the system of claim 6 said trained neural network to encode at least one of: image files, audio files, and large data set files[¶¶51-52] At a first step, an autoencoder 700 (exemplary) built from Convolutional Neural Network (CNN) discriminators 704, 706 using a CGAN generator 702 may be trained to extract the feature representations of the digit images. FIG. 7A shows the autoencoder 700 for performing integer addition over data. An adversarial training of the autoencoder 700 may be performed. As will be appreciated by those skilled in the art, a denoising autoencoder may include two CNNs (discriminators) with back to back pooling. It will be further appreciated that such an architecture may also be multi layered with multiple CNN and pooling layers. Further, a dense ANN part in the middle may also be multi-layered. A left half of the CGAN generator 702 may perform convolution and a right half of the CGAN generator 702 may perform image reconstruction from the feature vector. The feature vector may be an output from a dense layer in the middle. As will be further appreciated, the training of the CGAN generator 702 may be performed in two parts. The CGAN generator 702 may keep trying to trick the CNN discriminators 704, 706 to recognize fake images as real. The CNN discriminators 704, 706 may keep trying to recognize images created by the CGAN generator 702 created as fake. The training may continue until there is no progress in learning, i.e., the CGAN generator 702 generates the image replicas that are indistinguishable from the real image of the number. At this point, all weights of the autoencoder 700 are trained well enough to enable correct feature extraction. Applying this training algorithm to the whole data space, the autoencoder 700 may be fully trained. For example, if the addends data space is in integer set (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10), then the result of binary addition operation will be set of (0, 1, 2, 3, 4, 5 . . . 20)].
Regarding claim 6, Akerib disclose the system of claim I wherein said APU is inplemented on one of: SRAM, non-volatile, and non-destructive metnory [¶124, Memory computation device 100 may be formed of any suitable memory array, such as an SRAM, a non-volatile, a volatile, and a non-destructive array and may be formed into a plurality of bit line processors 114, each processing one bit of a word and each word being stored in a column of associative memory array 140].
Claim 6 is rejected under 35 U.S.C. 103 as being unpatentable over US Patent Application No. (US US20180341642) (filed in IDS 11/06/2025) issued to Akerib and in view of US Patent No. Application (US2021/0342700) issued to ANTIC(filed in IDS 11/06/2025), and in view of DAUB (US2021/0004864) (filed in IDS 11/06/2025),and further in view of (US2021/0090348) issued to Croxford(filed in IDS 11/06/2025).
Regarding claim 6, Akerib, ANTIC, and DAUB do not explicitly disclose, however, Croxford the system of claim 1 wherein said neural proxy hash encoder comprises a trained neural network comprising a plurality of layers to encode input data into feature sets [¶23…components in the secure environment may have access to certain storage, e.g., secure or “protected” memory regions, which are not accessible to components and systems outside of the secure environment. For example, components and devices performing non-secure operations can be prevented from accessing any storage, e.g., region of memory, which is defined as being secure],and [¶¶27-29, A neural network may be trained to perform object detection, image segmentation, sound/speech recognition etc. by processing sensor-originated data, for example to determine whether an object of a predetermined class of objects is present in the real-world environment represented by the sensor-originated data. Training the neural network in this way may generate one or more kernels associated with at least some of the layers (such as layers of the neural network other than the input and output layers). Hence, the output of the training may be a plurality of kernels associated with a predetermined neural network architecture (for example with different kernels being associated with different respective layers of a multi-layer neural network architecture). The kernel data may be considered to correspond to weight data representative of weights to be applied to image data, as each element of a kernel may be considered to correspond to a weight, respectively…After the training phase, the neural network (which may be referred to as a trained neural network) can be used to detect the presence of objects of a predetermined class of objects, e.g. in input images. This process may be referred to as “classification” or “inference”. Classification typically involves convolution of the kernels obtained during the training phase with portions of the sensor-originated input, e.g. image patches of an image input to the neural network, to generate a feature map. The feature map may then be processed using at least one fully connected layer, e.g. to classify the object; although other types of processing may be performed], and [¶55, In examples, at least one of the AR (Augmented Reality) data and the object recognition data is encrypted, e.g. when written to secure storage or transferred between components at the first trust level or within the secure environment. For example, an encryption scheme may be implemented such that the data is encrypted before being stored or transferred. The data may be encrypted by applying an encryption algorithm thereto. In some examples, a hash function (e.g. a cryptographic hash function) may be applied to the data. Encrypting and/or hashing the data may ensure that the data has not been tampered with when it is obtained, e.g. for further processing. Encryption and/or hashing may be applied to any data outputted from the secure environment, e.g. by the AR system, to similarly ensure that the outputted data (e.g. non-secure modified AR data) has not been modified prior to receipt by an external device such as a display device].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the teaching of Akerib, ANTIC, and DAUB with the teaching of Croxford in order to implement a neural network which may be trained to perform object detection, image segmentation, sound/speech recognition etc. The neural network (which may be referred to as a trained neural network) can be used to detect the presence of objects of a predetermined class of objects, e.g. in input images. Encrypting and/or hashing the data may ensure that the data has not been tampered with when it is obtained, e.g. for further processing. Encryption and/or hashing may be applied to any data output to ensure that the outputted data has not been modified [ Croxford, ¶¶27-29, 55].
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's
disclosure.
See submitted 892 for more relevant references.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SHAHRIAR ZARRINEH whose telephone number is (571)272-1207. The examiner can normally be reached Monday-Friday, 8:30am-5:30pm.
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, Jorge Ortiz-Criado can be reached at 571-272-7624. 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.
/SHAHRIAR ZARRINEH/Primary Examiner, Art Unit 2496
