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 .
Claim Rejections – 35 USC § 101
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.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1: Claims 11-19 are method claims. Claims 1-10, 20 are machine/system/product claims. Therefore, claims 1-20 are directed to either a process, machine, manufacture or composition of matter.
With respect to claim 1:
Step 2A – Prong 1:
…
…
…
… create a plurality of encoded vectorized outputs corresponding to the plurality of versions, wherein each of the plurality of encoded vectorized outputs comprises a one-dimensional vector that corresponds to a respective encoded vectorized output; (mental process – a person can manually create a plurality of encoded vectorized outputs corresponding to the plurality of versions with the assistance of a pen/paper.)
create a plurality of delta vectors, wherein each of the plurality of delta vectors corresponds to differences between successive versions of the file and may be used to obtain a previous version from a newer version; (mental process – a person can manually create a plurality of delta vectors corresponding to the plurality of versions with the assistance of a pen/paper.)
…
…
…
…
…
…
…
Step 2A – Prong 2: This judicial exception is not integrated into a practical application.
A computing platform comprising: at least one processor; (mere instructions to apply the exception using a generic computer component – processor applies exception.)
a communication interface communicatively coupled to the at least one processor; (mere instructions to apply the exception using a generic computer component – communication interface applies exception.)
and memory storing computer-readable instructions that, when executed by the at least one processor, cause the computing platform to: (mere instructions to apply the exception using a generic computer component – memory storing computer-readable instructions applies exception.)
train, based on a plurality of versions of a file, a convolutional neural network, wherein training the convolutional neural network configures the convolutional neural network to: … (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: High level recitation of training the machine learning engine to create a plurality of encoded vectorized outputs.);
…
and store, at a first repository, a newest encoded vectorized output; (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)).
receive, from a user device, a request to obtain a previous version of the file; (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)).
send, based on the request, one or more commands directing the first repository to send the newest encoded vectorized output to the computing platform, wherein sending the one or more commands directing the first repository to send the newest encoded vectorized output to the computing platform causes the first repository to send the newest encoded vectorized output to the computing platform; (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)).
receive, from the first repository, the newest encoded vectorized output; (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)).
input the newest encoded vectorized output into the convolutional neural network; (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)).
output, using the convolutional neural network, the previous version of the file from the newest encoded vectorized output using a corresponding delta vector of the plurality of delta vectors; (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)).
and send, to the user device, the previous version of the file and one or more commands directing the user device to display the previous version of the file, wherein sending the one or more commands directing the user device to display the previous version of the file causes the user device to display the previous version of the file. (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)).
Step 2B: The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
A computing platform comprising: at least one processor; (mere instructions to apply the exception using a generic computer component – processor applies exception.)
a communication interface communicatively coupled to the at least one processor; (mere instructions to apply the exception using a generic computer component – communication interface applies exception.)
and memory storing computer-readable instructions that, when executed by the at least one processor, cause the computing platform to: (mere instructions to apply the exception using a generic computer component – memory storing computer-readable instructions applies exception.)
train, based on a plurality of versions of a file, a convolutional neural network, wherein training the convolutional neural network configures the convolutional neural network to: … (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: High level recitation of training the machine learning engine to create a plurality of encoded vectorized outputs.);
…
and store, at a first repository, a newest encoded vectorized output; (MPEP 2106.05(d)(II) indicate that merely “Storing and retrieving information in memory” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim – the newest encoded vectorized output is merely stored at a first repository). Thereby, a conclusion that the claimed distribute step is well-understood, routine, conventional activity is supported under Berkheimer.)
receive, from a user device, a request to obtain a previous version of the file; (MPEP 2106.05(d)(II) indicate that merely “Receiving or transmitting data over a network, e.g., using the Internet to gather data” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim – the previous version of the file is merely received from the user device). Thereby, a conclusion that the claimed distribute step is well-understood, routine, conventional activity is supported under Berkheimer.)
send, based on the request, one or more commands directing the first repository to send the newest encoded vectorized output to the computing platform, wherein sending the one or more commands directing the first repository to send the newest encoded vectorized output to the computing platform causes the first repository to send the newest encoded vectorized output to the computing platform; (MPEP 2106.05(d)(II) indicate that merely “Receiving or transmitting data over a network, e.g., using the Internet to gather data” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim – the one or more commands directing the first repository to send the newest encoded vectorized output is merely transmitted to the to the computing platform). Thereby, a conclusion that the claimed distribute step is well-understood, routine, conventional activity is supported under Berkheimer.)
receive, from the first repository, the newest encoded vectorized output; (MPEP 2106.05(d)(II) indicate that merely “Receiving or transmitting data over a network, e.g., using the Internet to gather data” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim – the newest encoded vectorized output is merely received from the first repository). Thereby, a conclusion that the claimed distribute step is well-understood, routine, conventional activity is supported under Berkheimer.)
input the newest encoded vectorized output into the convolutional neural network; (MPEP 2106.05(d)(II) indicate that merely “Receiving or transmitting data over a network, e.g., using the Internet to gather data” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim – the newest encoded vectorized output is merely input to the convolutional neural network). Thereby, a conclusion that the claimed distribute step is well-understood, routine, conventional activity is supported under Berkheimer.)
output, using the convolutional neural network, the previous version of the file from the newest encoded vectorized output using a corresponding delta vector of the plurality of delta vectors; (MPEP 2106.05(d)(II) indicate that merely “Receiving or transmitting data over a network, e.g., using the Internet to gather data” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim – the previous version of the file is merely output using the convolutional neural network). Thereby, a conclusion that the claimed distribute step is well-understood, routine, conventional activity is supported under Berkheimer.)
and send, to the user device, the previous version of the file and one or more commands directing the user device to display the previous version of the file, wherein sending the one or more commands directing the user device to display the previous version of the file causes the user device to display the previous version of the file. (MPEP 2106.05(d)(II) indicate that merely “Receiving or transmitting data over a network, e.g., using the Internet to gather data” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim – the previous version of the file is merely transmitted to the user device). Thereby, a conclusion that the claimed distribute step is well-understood, routine, conventional activity is supported under Berkheimer.)
With respect to claim 2:
Step 2A – Prong 1:
…
…
create, using the convolutional neural network, a first encoded vectorized output corresponding to the first version and a second encoded vectorized output corresponding to the second version;
create, using the convolutional neural network, a first delta vector between the first encoded vectorized output and the second encoded vectorized output, wherein the first delta vector may be used to obtain the first version from the second version;
…
Step 2A – Prong 2: This judicial exception is not integrated into a practical application.
The computing platform of claim 1, wherein the training further configures the convolutional neural network to: receive, from a second repository, a first version of the file and a second version of the file, wherein the first version and the second version are successive versions of the file; (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)).
input, into the convolutional neural network, the first version and the second version; (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)).
…
…
and store, at the first repository, the second encoded vectorized output. (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)).
Step 2B: The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
The computing platform of claim 1, wherein the training further configures the convolutional neural network to: receive, from a second repository, a first version of the file and a second version of the file, wherein the first version and the second version are successive versions of the file; (MPEP 2106.05(d)(II) indicate that merely “Receiving or transmitting data over a network, e.g., using the Internet to gather data” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim – the a first version of the file and a second version of the file is merely transmitted from the second repository). Thereby, a conclusion that the claimed distribute step is well-understood, routine, conventional activity is supported under Berkheimer.)
input, into the convolutional neural network, the first version and the second version; (MPEP 2106.05(d)(II) indicate that merely “Receiving or transmitting data over a network, e.g., using the Internet to gather data” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim – the first version and the second version is merely input into the convolutional neural network). Thereby, a conclusion that the claimed distribute step is well-understood, routine, conventional activity is supported under Berkheimer.)
…
…
and store, at the first repository, the second encoded vectorized output. (MPEP 2106.05(d)(II) indicate that merely “Storing and retrieving information in memory” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim – the second encoded vectorized output is merely stored at a first repository). Thereby, a conclusion that the claimed distribute step is well-understood, routine, conventional activity is supported under Berkheimer.)
With respect to claim 3:
Step 2A – Prong 1:
The computing platform of claim 1, wherein the convolutional neural network further comprises a first layer and a second layer, wherein each of the first layer and the second layer reduces a dimensionality of each of the plurality of versions. (mental process – a person can recognize that the convolutional neural network further comprises a first layer and a second layer, wherein each of the first layer and the second layer reduces a dimensionality of each of the plurality of versions.)
With respect to claim 4:
Step 2A – Prong 1:
The computing platform of claim 1, wherein the creating the plurality of encoded vectorized outputs is performed iteratively until a reconstruction loss threshold is reached so that each of the plurality of encoded vectorized outputs may be used to output the corresponding plurality of versions. (mental process – a person can recognize that the creating the plurality of encoded vectorized outputs is performed iteratively until a reconstruction loss threshold is reached so that each of the plurality of encoded vectorized outputs may be used to output the corresponding plurality of versions.)
With respect to claim 5:
Step 2A – Prong 1:
The computing platform of claim 1, wherein the creating the plurality of delta vectors is performed iteratively until a reconstruction loss threshold is reached that enables each of the plurality of delta vectors to be used to obtain previous encoded vectorized outputs from successive encoded vectorized outputs. (mental process – a person can recognize that the creating the plurality of delta vectors is performed iteratively until a reconstruction loss threshold is reached that enables each of the plurality of delta vectors to be used to obtain previous encoded vectorized outputs from successive encoded vectorized outputs.)
With respect to claim 6:
Step 2A – Prong 1:
The computing platform of claim 1, (mental process from claim 1)
wherein the memory stores computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to: (mere instructions to apply the exception using a generic computer component – memory, processor applies exception.)
Step 2A – Prong 2: This judicial exception is not integrated into a practical application.
store the plurality of versions of the file at a second repository. (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)).
Step 2B: The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
store the plurality of versions of the file at a second repository. (MPEP 2106.05(d)(II) indicate that merely “Storing and retrieving information in memory” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim – the plurality of versions of the file is merely stored at a second repository). Thereby, a conclusion that the claimed distribute step is well-understood, routine, conventional activity is supported under Berkheimer.)
With respect to claim 7:
Step 2A – Prong 1:
The computing platform of claim 1, wherein the memory stores computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to: (mere instructions to apply the exception using a generic computer component – memory, processor applies exception.)
Step 2A – Prong 2: This judicial exception is not integrated into a practical application.
store the plurality of delta vectors at a bottleneck associated with the convolutional neural network. (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)).
Step 2B: The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
store the plurality of delta vectors at a bottleneck associated with the convolutional neural network. (MPEP 2106.05(d)(II) indicate that merely “Storing and retrieving information in memory” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim – the plurality of delta vectors is merely stored at a bottleneck associated with the convolutional neural network). Thereby, a conclusion that the claimed distribute step is well-understood, routine, conventional activity is supported under Berkheimer.)
With respect to claim 8:
Step 2A – Prong 1:
The computing platform of claim 1, wherein the memory stores computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to: pre-process the plurality of versions of the file before training the convolutional neural network, wherein the pre-processing comprises converting the plurality of versions of the file into a machine-readable format by tokenizing the plurality of versions of the file. (mental process – a person can manually pre-process the plurality of versions of the file by converting the plurality of versions of the file into a machine-readable format by tokenizing the plurality of versions of the file with the assistance of a pen/paper.)
With respect to claim 9:
Step 2A – Prong 1:
…
…
create, …, a third encoded vectorized output corresponding to the third version; (mental process – a person can manually create a third encoded vectorized output corresponding to the third version with the assistance of a pen/paper.)
and create, …, a second delta vector between the second encoded vectorized output and the third encoded vectorized output, wherein the second delta vector may be used to obtain the second version from the third version. (mental process – a person can manually create a second delta vector between the second encoded vectorized output and the third encoded vectorized output with the assistance of a pen/paper.)
Step 2A – Prong 2: This judicial exception is not integrated into a practical application.
The computing platform of claim 2, wherein the training further configures the convolutional neural network to: receive, from the second repository, a third version of the file, wherein the third version is a successive version of the second version of the file; (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)).
input, into the convolutional neural network, the third version; (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)).
… using the convolutional neural network … (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: High level recitation of training the machine learning engine to create a plurality of encoded vectorized outputs.);
… using the convolutional neural network … (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: High level recitation of training the machine learning engine to create a plurality of encoded vectorized outputs.);
Step 2B: The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
The computing platform of claim 2, wherein the training further configures the convolutional neural network to: receive, from the second repository, a third version of the file, wherein the third version is a successive version of the second version of the file; (MPEP 2106.05(d)(II) indicate that merely “Receiving or transmitting data over a network, e.g., using the Internet to gather data” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim – the third version of the file is merely received by the second repository). Thereby, a conclusion that the claimed distribute step is well-understood, routine, conventional activity is supported under Berkheimer.)
input, into the convolutional neural network, the third version; (MPEP 2106.05(d)(II) indicate that merely “Receiving or transmitting data over a network, e.g., using the Internet to gather data” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim – the third version is merely input into the convolutional neural network). Thereby, a conclusion that the claimed distribute step is well-understood, routine, conventional activity is supported under Berkheimer.)
… using the convolutional neural network … (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: High level recitation of training the machine learning engine to create a plurality of encoded vectorized outputs.);
… using the convolutional neural network … (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: High level recitation of training the machine learning engine to create a plurality of encoded vectorized outputs.);
With respect to claim 10:
Step 2A – Prong 1:
The computing platform of claim 9, wherein the training further configures the convolutional neural network to: (mental process from claim 9)
Step 2A – Prong 2: This judicial exception is not integrated into a practical application.
send the third encoded vectorized output and one or more commands directing the first repository to replace the second encoded vectorized output with the third encoded vectorized output, wherein sending the third encoded vectorized output and one or more commands directing the first repository to replace the second encoded vectorized output with the third encoded vectorized output causes the first repository to replace the second encoded vectorized output with the third encoded vectorized output. (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)).
Step 2B: The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
send the third encoded vectorized output and one or more commands directing the first repository to replace the second encoded vectorized output with the third encoded vectorized output, wherein sending the third encoded vectorized output and one or more commands directing the first repository to replace the second encoded vectorized output with the third encoded vectorized output causes the first repository to replace the second encoded vectorized output with the third encoded vectorized output. (MPEP 2106.05(d)(II) indicate that merely “Receiving or transmitting data over a network, e.g., using the Internet to gather data” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim – the third encoded vectorized output and one or more commands is merely transmitted to the first repository). Thereby, a conclusion that the claimed distribute step is well-understood, routine, conventional activity is supported under Berkheimer.)
Claims 11-15 are rejected on the same grounds under 35 U.S.C. 101 as claims 1-5 as they are substantially similar, respectively. Mutatis mutandis.
Claims 16-19 are rejected on the same grounds under 35 U.S.C. 101 as claims 7-10 as they are substantially similar, respectively. Mutatis mutandis.
Claim 20 is rejected on the same grounds under 35 U.S.C. 101 as claim 1 as they are substantially similar. Mutatis mutandis.
Claim Rejections – 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over Li et al. (US20240282136A1) hereinafter known as Li in view of Therrian et al. (US20040088331A1) hereinafter known as Therrian.
Regarding independent claim 1, Li teaches:
A computing platform comprising: at least one processor; (Li [¶ 0018]: “computer 101 includes processor set” Li teaches a computing platform with a processor.)
a communication interface communicatively coupled to the at least one processor; and memory storing computer-readable instructions that, when executed by the at least one processor, cause the computing platform to: (Li [¶ 0018]: “computer 101 includes processor set … computer 101 includes processor set 110 … volatile memory 112, persistent storage 113 … and network module” Li teaches a network module which is a communication interface as well as memory storing instructions.)
train, based on a plurality of versions of a file, a convolutional neural network, wherein training the convolutional neural network configures the convolutional neural network to: create a plurality of encoded vectorized outputs corresponding to the plurality of versions, wherein each of the plurality of encoded vectorized outputs comprises a one-dimensional vector that corresponds to a respective encoded vectorized output; (Li [¶ 0048]: “During training, sets of training documents are used that represent different versions of documents” Li teaches that sets of training documents, where the documents have different versions, are used to teach the neural network. Li [¶ 0045]: “Encoder 510 may include a convolutional neural network that processes matrix representation 505 to output a feature map 515 comprising a plurality of vectors” Li teaches that the neural network creates a feature map comprising vectors. Fig. 5 shows that the encoder encodes the versions of documents to the feature map.)
…
…
…
…
…
input the newest encoded vectorized output into the convolutional neural network; (Li [¶ 0047]: “decoder 550 is utilized to generate a reconstructed matrix representation 570” Li teaches that the encoded vector is fed into the decoder to generate a reconstructed matrix representation.)
…
and send, to the user device, the previous version of the file and one or more commands directing the user device to display the previous version of the file, wherein sending the one or more commands directing the user device to display the previous version of the file causes the user device to display the previous version of the file. (Li [¶ 0028]: “this recommendation would typically be communicated from network module 115 of computer 101 through WAN 102 to EUD 103. In this way, EUD 103 can display, or otherwise present, the recommendation to an end user” Li teaches that the output/recommendation, which is the previous version of the file, can be presented to the user device and displayed.)
Li does not explicitly teach:
create a plurality of delta vectors, wherein each of the plurality of delta vectors corresponds to differences between successive versions of the file and may be used to obtain a previous version from a newer version;
and store, at a first repository, a newest encoded vectorized output;
receive, from a user device, a request to obtain a previous version of the file;
send, based on the request, one or more commands directing the first repository to send the newest encoded vectorized output to the computing platform, wherein sending the one or more commands directing the first repository to send the newest encoded vectorized output to the computing platform causes the first repository to send the newest encoded vectorized output to the computing platform;
receive, from the first repository, the newest encoded vectorized output;
…
output, using the convolutional neural network, the previous version of the file from the newest encoded vectorized output using a corresponding delta vector of the plurality of delta vectors;
…
However, Therrian teaches:
create a plurality of delta vectors, wherein each of the plurality of delta vectors corresponds to differences between successive versions of the file and may be used to obtain a previous version from a newer version; (Therrian [¶ 0044]: “allows reverse delta compression to be applied to successive versions of a file” Therrian teaches the differences/delta between successive versions.)
and store, at a first repository, a newest encoded vectorized output; (Therrian [¶ 0044]: “It retains the latest version of the file in its uncompressed form and all earlier versions in reverse delta compressed form.” Therrian teaches that the differences/delta between the successive versions of the documents is stored.)
receive, from a user device, a request to obtain a previous version of the file; (Therrian [¶ 0043]: “a user might want to know when a particular change to a document was made or they may need to access an earlier version of a file for reference or modification.” Therrian teaches that a user may request a earlier version of the file for reference or modification.)
send, based on the request, one or more commands directing the first repository to send the newest encoded vectorized output to the computing platform, wherein sending the one or more commands directing the first repository to send the newest encoded vectorized output to the computing platform causes the first repository to send the newest encoded vectorized output to the computing platform; (Therrian [¶ 0044]: “To recreate an earlier version of a file, the reverse delta decompression algorithm starts with the latest version of the file and decompresses backward in a chain of versions until the desired file is reached” Therrian teaches that the reverse delta algorithm starts with the latest version of the file, which is the newest encoded vectorized output.)
receive, from the first repository, the newest encoded vectorized output; (Therrian [¶ 0044]: “To recreate an earlier version of a file, the reverse delta decompression algorithm starts with the latest version of the file and decompresses backward in a chain of versions until the desired file is reached” Therrian teaches that the reverse delta algorithm starts with the latest version of the file, which is the newest encoded vectorized output. The repository that runs the algorithm receives this file.)
…
output, using the convolutional neural network, the previous version of the file from the newest encoded vectorized output using a corresponding delta vector of the plurality of delta vectors; (Therrian [¶ 0044]: “To recreate an earlier version of a file, the reverse delta decompression algorithm starts with the latest version of the file and decompresses backward in a chain of versions until the desired file is reached” Therrian teaches that the reverse delta algorithm uses the newest version of the file to output a previous version of the file using the delta vectors.)
Li and Therrian are in the same field of endeavor as the present invention, as the references are directed to using versions of a file/document to train a neural network and using delta comparisons between successive versions of files, respectively. It would have been obvious, before the effective filing date of the claimed invention, to a person of ordinary skill in the art, to combine encoding a plurality of versions of a file/document as taught in Li with creating delta vectors between consecutive/successive versions as taught in Therrian. Therrian provides this additional functionality. As such, it would have been obvious to one of ordinary skill in the art to modify the teachings of Li to include teachings of Therrian because the combination would allow for older versions of a file to be efficiently found using a neural network. This has the potential benefit of speeding up accessing versions of files across many files in a large directory.
Regarding dependent claim 2, Li and Therrian teach:
The computing platform of claim 1, wherein the training further configures the convolutional neural network to: receive, from a second repository, a first version of the file and a second version of the file, wherein the first version and the second version are successive versions of the file; (Therrian [¶ 0044]: “allows reverse delta compression to be applied to successive versions of a file” Therrian teaches the differences/delta between successive versions. From a repository, two consecutive versions may be received.)
input, into the convolutional neural network, the first version and the second version; (Li [¶ 0048]: “During training, sets of training documents are used that represent different versions of documents” Li teaches that sets of training documents, where the documents have different versions, are used to teach the neural network.)
create, using the convolutional neural network, a first encoded vectorized output corresponding to the first version and a second encoded vectorized output corresponding to the second version; (Li [¶ 0045]: “Encoder 510 may include a convolutional neural network that processes matrix representation 505 to output a feature map 515 comprising a plurality of vectors” Li teaches that the neural network creates a feature map comprising vectors. Fig. 5 shows that the encoder encodes the versions of documents to the feature map.)
create, using the convolutional neural network, a first delta vector between the first encoded vectorized output and the second encoded vectorized output, wherein the first delta vector may be used to obtain the first version from the second version; (Therrian [¶ 0044]: “allows reverse delta compression to be applied to successive versions of a file” Therrian teaches the differences/delta between successive versions.)
and store, at the first repository, the second encoded vectorized output. (Therrian [¶ 0044]: “It retains the latest version of the file in its uncompressed form and all earlier versions in reverse delta compressed form.” Therrian teaches that the differences/delta between the successive versions of the documents is stored.)
The reasons to combine are substantially similar to those of claim 1.
Regarding dependent claim 3, Li and Therrian teach:
The computing platform of claim 1, wherein the convolutional neural network further comprises a first layer and a second layer, wherein each of the first layer and the second layer reduces a dimensionality of each of the plurality of versions. (Li [¶ 0045]: “In the depicted example, matrix representation 505 is a 1024×1024 matrix, and feature map 515 is a 64×64×256 embedding space” Li teaches that the first layer and the second layer reduces the dimensionality, namely from a 1024x1024 matrix to a feature map of 64x64x256 embedding space.)
The reasons to combine are substantially similar to those of claim 1.
Regarding dependent claim 4, Li and Therrian teach:
The computing platform of claim 1, wherein the creating the plurality of encoded vectorized outputs is performed iteratively until a reconstruction loss threshold is reached so that each of the plurality of encoded vectorized outputs may be used to output the corresponding plurality of versions. (“Li [¶ 0058]: “the embedding spaces of the encoder, decoder, and codebook are adjusted using a loss function so that the reconstructed matrix representation of each training document most closely resembles the original matrix representation of that training document” Li teaches that the document is adjusted until (showing iterative approach) the loss function reaches the threshold of most closely resembling the original.)
The reasons to combine are substantially similar to those of claim 1.
Regarding dependent claim 5, Li and Therrian teach:
The computing platform of claim 1, wherein the creating the plurality of delta vectors is performed iteratively until a reconstruction loss threshold is reached that enables each of the plurality of delta vectors to be used to obtain previous encoded vectorized outputs from successive encoded vectorized outputs. (Therrian [¶ 0044]: “To recreate an earlier version of a file, the reverse delta decompression algorithm starts with the latest version of the file and decompresses backward in a chain of versions until the desired file is reached” Therrian teaches that the reverse delta algorithm uses the newest version of the file to output a previous version of the file using the delta vectors.)
The reasons to combine are substantially similar to those of claim 1.
Regarding dependent claim 6, Li and Therrian teach:
The computing platform of claim 1, wherein the memory stores computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to: store the plurality of versions of the file at a second repository. (Therrian [¶ 0044]: “To recreate an earlier version of a file, the reverse delta decompression algorithm starts with the latest version of the file and decompresses backward in a chain of versions until the desired file is reached” Therrian teaches that the reverse delta algorithm starts with the latest version of the file, which is the newest encoded vectorized output. The repository that runs the algorithm stores/receives this file.)
The reasons to combine are substantially similar to those of claim 1.
Regarding dependent claim 7, Li and Therrian teach:
The computing platform of claim 1, wherein the memory stores computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to: store the plurality of delta vectors at a bottleneck associated with the convolutional neural network. (Therrian [¶ 0044]: “To recreate an earlier version of a file, the reverse delta decompression algorithm starts with the latest version of the file and decompresses backward in a chain of versions until the desired file is reached” Therrian teaches that the reverse delta algorithm starts with the latest version of the file, which is the newest encoded vectorized output. The repository that runs the algorithm stores/receives this file.)
The reasons to combine are substantially similar to those of claim 1.
Regarding dependent claim 8, Li and Therrian teach:
The computing platform of claim 1, wherein the memory stores computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to: pre-process the plurality of versions of the file before training the convolutional neural network, wherein the pre-processing comprises converting the plurality of versions of the file into a machine-readable format by tokenizing the plurality of versions of the file. (Li [¶ 0045]: “Encoder 510 may include a convolutional neural network that processes matrix representation 505 to output a feature map 515 comprising a plurality of vectors” Li teaches that the neural network creates a feature map comprising vectors. Fig. 5 shows that the encoder encodes the versions of documents to the feature map.)
The reasons to combine are substantially similar to those of claim 1.
Regarding dependent claim 9, Li and Therrian teach:
The computing platform of claim 2, wherein the training further configures the convolutional neural network to: receive, from the second repository, a third version of the file, wherein the third version is a successive version of the second version of the file; (Therrian [¶ 0044]: “allows reverse delta compression to be applied to successive versions of a file” Therrian teaches the differences/delta between successive versions. From a repository, a third version of the file may be received.)
input, into the convolutional neural network, the third version; (Li [¶ 0047]: “decoder 550 is utilized to generate a reconstructed matrix representation 570” Li teaches that the encoded vector is fed into the decoder to generate a reconstructed matrix representation.)
create, using the convolutional neural network, a third encoded vectorized output corresponding to the third version; (Li [¶ 0045]: “Encoder 510 may include a convolutional neural network that processes matrix representation 505 to output a feature map 515 comprising a plurality of vectors” Li teaches that the neural network creates a feature map comprising vectors. Fig. 5 shows that the encoder encodes the versions of documents to the feature map.)
and create, using the convolutional neural network, a second delta vector between the second encoded vectorized output and the third encoded vectorized output, wherein the second delta vector may be used to obtain the second version from the third version. (Therrian [¶ 0044]: “To recreate an earlier version of a file, the reverse delta decompression algorithm starts with the latest version of the file and decompresses backward in a chain of versions until the desired file is reached” Therrian teaches that the reverse delta algorithm starts with the latest version of the file, which is the newest encoded vectorized output.)
The reasons to combine are substantially similar to those of claim 1.
Regarding dependent claim 10, Li and Therrian teach:
The computing platform of claim 9, wherein the training further configures the convolutional neural network to: send the third encoded vectorized output and one or more commands directing the first repository to replace the second encoded vectorized output with the third encoded vectorized output, wherein sending the third encoded vectorized output and one or more commands directing the first repository to replace the second encoded vectorized output with the third encoded vectorized output causes the first repository to replace the second encoded vectorized output with the third encoded vectorized output. (Li [¶ 0045]: “Encoder 510 may include a convolutional neural network that processes matrix representation 505 to output a feature map 515 comprising a plurality of vectors” Li teaches that the neural network creates a feature map comprising vectors. Fig. 5 shows that the encoder encodes the versions of documents to the feature map.)
The reasons to combine are substantially similar to those of claim 1.
Claims 11-15 are rejected on the same grounds under 35 U.S.C. 103 as claims 1-5 as they are substantially similar, respectively. Mutatis mutandis.
Claims 16-19 are rejected on the same grounds under 35 U.S.C. 103 as claims 7-10 as they are substantially similar, respectively. Mutatis mutandis.
Claim 20 is rejected on the same grounds under 35 U.S.C. 103 as claim 1 as they are substantially similar. Mutatis mutandis.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to KYU HYUNG HAN whose telephone number is (703) 756-5529. The examiner can normally be reached on MF 9-5.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Alexey Shmatov can be reached on (571) 270-3428. 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.
/Kyu Hyung Han/
Examiner
Art Unit 2123
/ALEXEY SHMATOV/Supervisory Patent Examiner, Art Unit 2123
Prosecution Insights