Prosecution Insights
Last updated: November 04, 2026
Application No. 18/416,704

LARGE LANGUAGE MODELS FOR EFFICIENT ANOMALY DETECTION IN LOG FILES OF INDUSTRIAL MACHINES

Non-Final OA §101§103§112
Filed
Jan 18, 2024
Examiner
NGUYEN, HENRY K
Art Unit
2121
Tech Center
2100 — Computer Architecture & Software
Assignee
Nano Dimension Technologies Ltd.
OA Round
1 (Non-Final)
60%
Grant Probability
Moderate
1-2
OA Rounds
1y 8m
Est. Remaining
86%
With Interview

Examiner Intelligence

Grants 60% of resolved cases
60%
Career Allowance Rate
100 granted / 168 resolved
+4.5% vs TC avg
Strong +27% interview lift
Without
With
+26.8%
Interview Lift
resolved cases with interview
Typical timeline
4y 5m
Avg Prosecution
15 currently pending
Career history
196
Total Applications
across all art units

Statute-Specific Performance

§101
21.2%
-18.8% vs TC avg
§103
53.1%
+13.1% vs TC avg
§102
7.9%
-32.1% vs TC avg
§112
13.0%
-27.0% vs TC avg
Black line = Tech Center average estimate • Based on career data from 168 resolved cases

Office Action

§101 §103 §112
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 . Claim Interpretation The limitation “predicting abnormal behavior in the one or more components of the industrial machine logged by the one or more loggers at the one or more subsequent times when the anomaly severity histogram indicates a pattern of the M probabilities associated with the abnormal behavior” of claim 1 is interpreted as a contingent limitation. The predicting step is only performed when the anomaly severity histogram indicates patterns associated with abnormal behavior and is not required by the claims. See MPEP 2111.04. Claim Rejections - 35 USC § 112 .The following is a quotation of 35 U.S.C. 112(b): (b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention. The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph: The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention. Claims 1-20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention. Claims 1, 10, and 19 recite the limitation “outputting from the large language model, based on the anomaly severity sequence at the one or more initial times, an anomaly severity histogram predicting a plurality of M distinct probabilities that the one or more loggers will record log file messages encoded by one or more log keys associated with M respective distinct anomaly severity levels at the one or more subsequent times”. It is unclear whether the loggers will record file messages encoded by one or more log keys associated with distinct severity levels at one or more subsequent times or not. For examination purposes, Examiner interprets the limitation the one or more loggers will record log file messages encoded by one or more log keys associated with M respective distinct anomaly severity levels at the one or more subsequent times. Claims 2-9, 11-18, and 20 are dependent claims that do not cure the deficiencies and are rejected for the same reasons. 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 1-6, 9-15, and 18-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. Step 1 According to the first part of the analysis, in the instant case, claims 1-9 are directed to a method, claims 10-18 are directed to a system comprising at least a processor, and claims 19-20 are directed to a non-transitory computer-readable medium. Thus, each of the claims falls within one of the four statutory categories (i.e. process, machine, manufacture, or composition of matter). Claim 1 recites: Step 2A, Prong 1 “transforming the sequence of log file messages into an anomaly severity sequence comprising a sequence of log tokens encoding the log file messages, time between log file token pairs, and a sequence of anomaly severity levels each associated with one or more of the log tokens” (This step is a recitation of a mental process that is practical to perform in the human mind. A human can convert a sequence of log file messages to a sequence of log tokens associated with a time and severity level in with the aid of paper and pen. (i.e., observation, evaluation, judgement, opinion). See MPEP § 2106.04(a)(2), subsection III.). “predicting abnormal behavior in the one or more components of the industrial machine logged by the one or more loggers at the one or more subsequent times when the anomaly severity histogram indicates a pattern of the M probabilities associated with the abnormal behavior” (This step is a recitation of a mental process that is practical to perform in the human mind. A human can predict abnormal behavior based on the anomaly severity indicated by a histogram. (i.e., observation, evaluation, judgement, opinion). See MPEP § 2106.04(a)(2), subsection III.). Step 2A, Prong 2 “receiving, from one or more loggers recording operations of one or more components in an industrial machine, a sequence of log file messages recorded by the one or more loggers at one or more initial times” (insignificant extra-solution activity) “inputting the anomaly severity sequence at the one or more initial times into the large language model trained to predict anomaly severity histograms at one or more subsequent times” (predicting an anomaly severity is a mental process. Using a LLM to predict the anomaly severity is mere instructions to apply the exception using a generic computer component. See 2106.05(f).) “outputting from the large language model, based on the anomaly severity sequence at the one or more initial times, an anomaly severity histogram predicting a plurality of M distinct probabilities that the one or more loggers will record log file messages encoded by one or more log keys associated with M respective distinct anomaly severity levels at the one or more subsequent times” (This step is directed to presenting information, is understood to be insignificant extra- solution activity as presenting offer and gathering statistics. See MPEP 2106.05(g).) Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application. Step 2B “receiving, from one or more loggers recording operations of one or more components in an industrial machine, a sequence of log file messages recorded by the one or more loggers at one or more initial times” (This step appears to be directed to transmitting or receiving information, which is well-understood, routine, and conventional. i. Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); See MPEP 2106.05 (d) (II).) “inputting the anomaly severity sequence at the one or more initial times into the large language model trained to predict anomaly severity histograms at one or more subsequent times” (predicting an anomaly severity is a mental process. Using a LLM to predict the anomaly severity is mere instructions to apply the exception using a generic computer component. See 2106.05(f).) “outputting from the large language model, based on the anomaly severity sequence at the one or more initial times, an anomaly severity histogram predicting a plurality of M distinct probabilities that the one or more loggers will record log file messages encoded by one or more log keys associated with M respective distinct anomaly severity levels at the one or more subsequent times” (This step is directed to presenting information, which is understood to be insignificant extra-solution activity and is well understood, routine and conventional activity of presenting offers as identified by the court (MPEP 2106.05(d)(ll)(iv))))) Even when considered in combination, these additional elements represent mere instructions to implement an abstract idea or other exception on a computer and insignificant extra-solution activity, which do not provide an inventive concept. Claim 2 recites: Step 2A, Prong 1 Claim 2 recites at least the abstract idea identified above in claim 1. Step 2A, Prong 2 “upon predicting the abnormal behavior, sending the industrial machine a control command to trigger the industrial machine to automatically execute an action to prevent the abnormal behavior from occurring before the one or more subsequent times.” (insignificant extra-solution activity) Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application. Step 2B “upon predicting the abnormal behavior, sending the industrial machine a control command to trigger the industrial machine to automatically execute an action to prevent the abnormal behavior from occurring before the one or more subsequent times.” (This step appears to be directed to transmitting or receiving information, which is well-understood, routine, and conventional. i. Receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 (utilizing an intermediary computer to forward information); See MPEP 2106.05 (d) (II).) Even when considered in combination, these additional elements represent mere instructions to implement an abstract idea or other exception on a computer and insignificant extra-solution activity, which do not provide an inventive concept. Claim 3 recites: Step 2A, Prong 1 Claim 3 recites at least the abstract idea identified above in claim 1. Step 2A, Prong 2 “wherein the control command triggers the industrial machine to automatically alter the operation of the one or more components predicted to have abnormal behavior” (linking judicial exception to a field of use. See MPEP 2106.05(h).) Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application. Step 2B “wherein the control command triggers the industrial machine to automatically alter the operation of the one or more components predicted to have abnormal behavior” (linking judicial exception to a field of use. See MPEP 2106.05(h).) Even when considered in combination, these additional elements represent mere instructions to implement an abstract idea or other exception on a computer and insignificant extra-solution activity, which do not provide an inventive concept. Claim 4 recites: Step 2A, Prong 1 “wherein the number M of anomaly severity levels is significantly smaller than a number N of log tokens” (This step is a recitation of a mental process that is practical to perform in the human mind. A human can determine that the number of anomaly severity levels is smaller than the number of log tokens. (i.e., observation, evaluation, judgement, opinion). See MPEP § 2106.04(a)(2), subsection III.). Step 2A, Prong 2 & 2B The claim does not recite any additional elements Claim 5 recites: Step 2A, Prong 1 “process the anomaly severity sequence that is represented by a data structure having a dimension on an order of a product of the number M of anomaly severity levels and a number of patterns representing interrelationships between log tokens and the anomaly severity levels” (This step is a recitation of a mental process that is practical to perform in the human mind. A human can process the anomaly severity sequence data (i.e., observation, evaluation, judgement, opinion). See MPEP § 2106.04(a)(2), subsection III.). Step 2A, Prong 2 “wherein one or more layers of the large language model process the anomaly severity sequence that is represented by a data structure having a dimension on an order of a product of the number M of anomaly severity levels and a number of patterns representing interrelationships between log tokens and the anomaly severity levels” (mere instructions to apply the exception using a generic computer component. See 2106.05(f).) Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application. Step 2B “wherein one or more layers of the large language model process the anomaly severity sequence that is represented by a data structure having a dimension on an order of a product of the number M of anomaly severity levels and a number of patterns representing interrelationships between log tokens and the anomaly severity levels” (mere instructions to apply the exception using a generic computer component. See 2106.05(f).) Even when considered in combination, these additional elements represent mere instructions to implement an abstract idea or other exception on a computer and insignificant extra-solution activity, which do not provide an inventive concept. Claim 6 recites: Step 2A, Prong 1 “wherein the anomaly severity histogram predicts the same number M of probabilities for any timescale defined by the one or more subsequent times” (This step is a recitation of a mental process that is practical to perform in the human mind. A human can determine that the number of anomaly severity levels is smaller than the number of log tokens. (i.e., observation, evaluation, judgement, opinion). See MPEP § 2106.04(a)(2), subsection III.). Step 2A, Prong 2 & 2B The claim does not recite any additional elements. Claim 9 recites: Step 2A, Prong 1 “label the sequence of log tokens with the sequence of anomaly severity levels” (This step is a recitation of a mental process that is practical to perform in the human mind. A human can label a sequence of log tokens with severity levels (i.e., observation, evaluation, judgement, opinion). See MPEP § 2106.04(a)(2), subsection III.). Step 2A, Prong 2 “performing self-supervised training to automatically label the sequence of log tokens with the sequence of anomaly severity levels” (mere instructions to apply the exception using a generic computer component. See 2106.05(f).) Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application. Step 2B “performing self-supervised training to automatically label the sequence of log tokens with the sequence of anomaly severity levels” (mere instructions to apply the exception using a generic computer component. See 2106.05(f).) Even when considered in combination, these additional elements represent mere instructions to implement an abstract idea or other exception on a computer and insignificant extra-solution activity, which do not provide an inventive concept. Claim 10 recites: See rejection of claim 1. Same rationale applies. Step 2A, Prong 2 & 2B The claim recites additional elements (“a system”, “memories”, and “processor”). (Mere instructions to apply the exception using a generic computer component. See 2106.05(f).) This judicial exception is not integrated into a practical application. The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. Claim 11 recites: Step 2A, Prong 1 See rejection of claim 2. Same rationale applies. Claim 12 recites: Step 2A, Prong 1 See rejection of claim 3. Same rationale applies. Claim 13 recites: Step 2A, Prong 1 See rejection of claim 4. Same rationale applies. Claim 14 recites: Step 2A, Prong 1 See rejection of claim 5. Same rationale applies. Claim 15 recites: Step 2A, Prong 1 See rejection of claim 6. Same rationale applies. Claim 18 recites: Step 2A, Prong 1 See rejection of claim 9. Same rationale applies. Claim 19 recites: See rejection of claim 1. Same rationale applies. Step 2A, Prong 2 & 2B The claim recites additional elements (“a system”, “memories”, and “processor”). (Mere instructions to apply the exception using a generic computer component. See 2106.05(f).) This judicial exception is not integrated into a practical application. The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. Claim 20 recites: Step 2A, Prong 1 See rejection of claim 2. Same rationale applies. 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 (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status. The following is a quotation of 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, 5-10, and 14-20 are rejected under 35 U.S.C. 103 as being unpatentable over Du et al. (“DeepLog: Anomaly Detection and Diagnosis from System Logs through Deep Learning”) in view of Zhang et al. (“Cat: Beyond efficient transformer for content-aware anomaly detection in event sequences”). Regarding Claim 1, Du (“DeepLog: Anomaly Detection and Diagnosis from System Logs through Deep Learning”) teaches a method of detecting abnormal behavior in industrial machines, the method comprising: receiving, from one or more loggers recording operations of one or more components in an industrial machine (pg. 1285; “System logs record system states and signi cant events at various critical points to help debug performance issues and failures, and perform root cause analysis. Such log data is universally available in nearly all computer systems and is a valuable resource for un derstanding system status.”), a sequence of log file messages recorded by the one or more loggers at one or more initial times; transforming the sequence of log file messages into an anomaly severity sequence comprising a sequence of log tokens encoding the log file messages (pg.1286, section 2.1; “As shown by several prior work [9, 22, 39, 42, 45], an e ective methodology is to extract a “log key” (also known as “message type”) from each log entry. e log key of a log entry e refers to the string constantk from the print statement in the source code which printede during the execution of that code.” Log key (i.e., log token).), time between log file token pairs (pg. 1286, section 2.1; “Past work on log analysis [22, 39, 42, 44] have discarded times tamp and/or parameter values in a log entry, and only used log keys to detect anomalies. DeepLog stores parameter values for each log → entry e, as well as the time elapsed between e and its predecessor, into a vector − v e.”), and a sequence of anomaly severity levels each associated with one or more of the log tokens (pg. 1287; “We train DeepLog as a multi-class classi er over recent context. e input is a history of recent log keys, and the output is a probability distribution over the n log keys from K, representing the probability that the next log key in the sequence is a key ki ∈ K.” probability (i.e., severity level).); outputting from the large language model, based on the anomaly severity sequence at the one or more initial times, predicting a plurality of M distinct probabilities that the one or more loggers will record log file messages encoded by one or more log keys associated with M respective distinct anomaly severity levels at the one or more subsequent times (pg. 1287, section 3.1; “We can model anomaly detection in a log key sequence as a multi-class classification problem, where each distinct log key defines a class. We train DeepLog as a multi-class classifier over recent context. e input is a history of recent log keys, and the output is a probability distribution over the n log keys from K, representing the probability that the next log key in the sequence is a key ki ∈ K.”); and Du does not explicitly disclose inputting the anomaly severity sequence at the one or more initial times into the large language model trained to predict anomaly severity histograms at one or more subsequent times; an anomaly severity histogram predicting a plurality of M distinct probabilities that the one or more loggers will record log file messages encoded by one or more log keys associated with M respective distinct anomaly severity levels at the one or more subsequent times However, Zhang (“CAT: Beyond Efficient Transformer for Content-Aware Anomaly Detection in Event Sequences”) teaches inputting the anomaly severity sequence at the one or more initial times into the large language model trained to predict anomaly severity histograms at one or more subsequent times (pg. 4542; “CAT adopts a self-attentive encoder-decoder transformer architecture, where the encoder learns event sequence representations, and the decoder embeds event sequences into a latent space, where anomalies are distinguishable. Specifically, we first design a content-awareness layer to process the event content information, generating representations of each event. The event representations are then grouped as event representation sequences as the input of the encoder and decoder.” Log sequences inputted into LLM. Pg. 4543, section 3; “Given a set of event sequences S = {𝑆1,𝑆2,...,𝑆|S|} as the training set, An anomaly detection model learns key patterns that given a new event sequence, the model could indicate if it is a normal sequence or an anomaly.” Pg. 4548; “Figure 6: Histogram of sequence lengths and log template indices (alphabets) distribution of normal and anomaly log sequences in Thunderbird dataset.”); an anomaly severity histogram predicting a plurality of M distinct probabilities that the one or more loggers will record log file messages encoded by one or more log keys associated with M respective distinct anomaly severity levels at the one or more subsequent times (pg. 4548, section 5.4; “. Figure 6 shows the his togram of sequence lengths and log template indices (alphabets) distribution (doughnut charts) of normal and anomaly sequences. From the histogram we can see that the sequence length distribu tion of normal and anomaly sequences differs a lot: The 5% and 95% percentile of normal log sequence length is 30 and 667, while for anomaly sequences they are 138 and 1484.”) Du and Zhang are analogous because they are directed to the field of log based anomaly detection using language models. It would have been obvious to one of ordinary skill to modify the anomaly detection model of Du with the histogram of Zhang. Doing so would allow for visualizing differences between the sequence length distribution of a normal sequence and the anomalous sequence (Zhang pg. 4548). Regarding Claim 5, Du and Zhang teach the method of claim 1. Du further teaches wherein one or more layers of the large language model process the anomaly severity sequence that is represented by a data structure having a dimension on an order of a product of the number M of anomaly severity levels and a number of patterns representing interrelationships between log tokens and the anomaly severity levels (pg. 1289, section 3.1; “The input layer encodes the n possible log keys from K as one-hot vectors. at is, a sparse n-dimensional vector − → u i is constructed for the log key ki ∈ K, such that − → u i[i] = 1 and −→ ui[j] = 0 for all other j i.”). Regarding Claim 6, Du and Zhang teach the method of claim 1, Zhang further teaches wherein the anomaly severity histogram predicts the same number M of probabilities for any timescale defined by the one or more subsequent times (pg. 4548, section 5.4; “From the histogram we can see that the sequence length distribution of normal and anomaly sequences differs a lot: The 5% and 95% percentile of normal log sequence length is 30 and 667, while for anomaly sequences they are 138 and 1484. Such long-sequence pat terns cannot be captured under a short fixed observation window, which may be the reason for the failure of observation window based approaches in the Thunderbird dataset.”). Du and Zhang are analogous because they are directed to the field of log based anomaly detection using language models. It would have been obvious to one of ordinary skill to modify the anomaly detection model of Du with the histogram of Zhang. Doing so would allow for visualizing differences between the sequence length distribution of a normal sequence and the anomalous sequence (Zhang pg. 4548). Regarding Claim 7, Du and Zhang teach the method of claim 1. Du further teaches wherein the large language model is a single- sequence transformer, the method further comprising: generating embedded input and/or output severity vectors representing interrelationships between entries of input and/or output anomaly severity sequence (pg. 1289, section 3.2; “Consider the example in Table 1. e time series for the parameter value vector sequence ofk2 is: {[t2−t1,0.61],[t2−t1,1]}. Hence, our problem is reduced to anomaly detection from a multi-variate time series data.”), wherein the embedded input and/or output severity vectors have a dimension on an order of a product of embedding input and/or output severity vectors (pg. 1289, section 3.1; “The input layer encodes the n possible log keys from K as one-hot vectors. at is, a sparse n-dimensional vector − → u i is constructed for the log key ki ∈ K, such that − → u i[i] = 1 and −→ ui[j] = 0 for all other j i.”); and processing the embedded input and/or output severity vectors at an encoder and/or decoder of the transformer (pg. 1289, section 3.1; “For simplicity, it omits an input layer and an output layer constructed by standard encoding-decoding schemes. The input layer encodes the n possible log keys from K as one-hot vectors.”). Regarding Claim 8, Du and Zhang teach the method of claim 1. Zhang further teaches wherein the large language model is a multi- sequence transformer, the method further comprising: inputting a plurality of anomaly severity sequences of log file tokens transformed from a plurality of sequences of log file messages generated by a plurality of respective loggers (pg. 4543, section 3; “Event: An event is considered as a sequence of tokens. Formally, an event 𝑒 = {𝑡𝑜𝑘1,𝑡𝑜𝑘2,...,𝑡𝑜𝑘|𝑒|}, where 𝑡𝑜𝑘𝑖 represents the 𝑖-th tokenized words/symbols or numbers, |𝑒| is the total number of tokens.”); generating a plurality of embedded input and/or output severity vectors derived from the plurality of respective anomaly severity sequences (pg. 4542; “However, they use simple aggregation of each event content token’s representation as the event embedding.”); inputting the plurality of embedded input and/or output severity vectors into a plurality of distinct respective intra sequence multi-head self-attention layers to output a plurality of distinct respective sets of intra sequence attention vectors (figure 3; pg. 4545; “For model stability concern, a multi-head attention mechanism is usually applied: MultiHead(𝑄,𝐾,𝑉) = Concat(head1,head2,...,headℎ)𝑊𝑂 where head𝑖 = Attention(𝑄𝑊𝑄 𝑖 ,𝐾𝑊𝐾 𝑖 ,𝑉𝑊𝑉 𝑖 ) Here 𝑊𝑄 𝑖 ,𝑊𝐾 𝑖 ,𝑊𝑉 𝑖 are projection matrices for each head 𝑖, 𝑊𝑂 denotes to the projection matrix of the concatenated multi-heads.”), each set identifying patterns associated with relationships among entries within the same anomaly severity sequence associated with the same logger (pg. 4542; “In this paper, we propose to approach the problem by introducing a self-attentive encoder-decoder transformer model to capture both sequential pattern and semantic information in event sequences.”); and inputting embedding patterns derived from a combination of the plurality of anomaly severity sequences into a same inter sequence multi-head self-attention layer to output a plurality of interrelated sets of inter sequence attention vectors identifying patterns associated with relationships among entries across multiple different ones of the anomaly severity sequences generated by multiple different loggers (figure 3; pg. 4545, section 4.2; “denotes to the projection matrix of the concatenated multi-heads. Each of the multi-head attention layer is followed by a fully con nected feed-forward network: FFN(𝑥) = ReLU(𝑥𝑊1 +𝑏1)𝑊2 +𝑏2 In the decoder part, a step-by-step dynamic decoding process is usually applied that predicts output representations from 𝑘-th to (𝑘 +1)-th step, until reaching the end of the sequence.”). Du and Zhang are analogous because they are directed to the field of log based anomaly detection using language models. It would have been obvious to one of ordinary skill to modify the anomaly detection model of Du with the histogram of Zhang. Doing so would allow for visualizing differences between the sequence length distribution of a normal sequence and the anomalous sequence (Zhang pg. 4548). Regarding Claim 9, Du and Zhang teach the method of claim 1. Du further teaches comprising performing self-supervised training to automatically label the sequence of log tokens with the sequence of anomaly severity levels (pg. 1289, section 3.1; “training step entails ending proper assignments to the weights so that the nal output of the sequence of LSTMs produces the de sired label (output) that comes with inputs in the training data set.”). Regarding Claim 10, Claim 10 is the system corresponding to the method of claim 1. Claim 10 is substantially similar to claim 1 and is rejected on the same grounds. Regarding Claim 14, Claim 14 is the system corresponding to the method of claim 5. Claim 14 is substantially similar to claim 5 and is rejected on the same grounds. Regarding Claim 15, Claim 15 is the system corresponding to the method of claim 6. Claim 15 is substantially similar to claim 6 and is rejected on the same grounds. Regarding Claim 16, Claim 16 is the system corresponding to the method of claim 7. Claim 16 is substantially similar to claim 7 and is rejected on the same grounds. Regarding Claim 17, Claim 17 is the system corresponding to the method of claim 8. Claim 17 is substantially similar to claim 8 and is rejected on the same grounds. Regarding Claim 18, Claim 18 is the system corresponding to the method of claim 9. Claim 18 is substantially similar to claim 9 and is rejected on the same grounds. Regarding Claim 19, Claim 19 is the computer-readable storage medium corresponding to the method of claim 1. Claim 19 is substantially similar to claim 1 and is rejected on the same grounds. Regarding Claim 20, Claim 20 is the computer-readable storage medium corresponding to the method of claim 1. Claim 20 is substantially similar to claim 2 and is rejected on the same grounds. Claims 2-3 and 11-12 are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Du/Zhang, as applied above, and further in view of Raymond et al. (US-20250132817-A1). Regarding Claim 2, Du and Zhang teach the method of claim 1. Du and Zhang do not explicitly disclose comprising, upon predicting the abnormal behavior, sending the industrial machine a control command to trigger the industrial machine to automatically execute an action to prevent the abnormal behavior from occurring before the one or more subsequent times. However, Raymond (US 20250132817 A1) teaches comprising, upon predicting the abnormal behavior, sending the industrial machine a control command to trigger the industrial machine to automatically execute an action to prevent the abnormal behavior from occurring before the one or more subsequent times (para [0046] “Furthermore, some embodiments of the satellite signal optimization system 100 include an artificial intelligence (or machine learning) engine that predicts space segment anomalies (e.g., signal throughput, satellite velocity, satellite position, satellite orientation, and the like) and generates configuration changes to the satellite spacecraft 120 that mitigate or correct the space segment anomalies (e.g., signal throughput, satellite velocity, satellite position, satellite orientation, and the like).”). Du, Zhang, and Raymond are analogous because they are directed towards super intelligence anomaly detection. It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the anomaly detection system of Du and Zhang with the action mitigation of Raymond. Doing so would allow for generation configurations to mitigate or correct the anomalies (Raymond para [0046]). Regarding Claim 3, Du, Zhang, and Raymond teach the method of claim 2. Raymond further teaches wherein the control command triggers the industrial machine to automatically alter the operation of the one or more components predicted to have abnormal behavior (para [0046] “Furthermore, some embodiments of the satellite signal optimization system 100 include an artificial intelligence (or machine learning) engine that predicts space segment anomalies (e.g., signal throughput, satellite velocity, satellite position, satellite orientation, and the like) and generates configuration changes to the satellite spacecraft 120 that mitigate or correct the space segment anomalies (e.g., signal throughput, satellite velocity, satellite position, satellite orientation, and the like).”). Du, Zhang, and Raymond are analogous because they are directed towards super intelligence anomaly detection. It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the anomaly detection system of Du and Zhang with the action mitigation of Raymond. Doing so would allow for generation configurations to mitigate or correct the anomalies (Raymond para [0046]). Regarding Claim 11, Claim 11 is the system corresponding to the method of claim 2. Claim 11 is substantially similar to claim 2 and is rejected on the same grounds. Regarding Claim 12, Claim 12 is the system corresponding to the method of claim 3. Claim 12 is substantially similar to claim 3 and is rejected on the same grounds. Claims 4 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Du/Zhang, as applied above, and further in view of Bai et al. (US-20250028737-A1). Regarding Claim 4, Du and Zhang teach the method of claim 1. Du and Zhang do not explicitly disclose wherein the number M of anomaly severity levels is significantly smaller than a number N of log tokens. However, Bai (US 20250028737 A1) teaches wherein the number M of anomaly severity levels (para [0092] “For example, the anomaly boundary corresponding to the variation range of history normal behavior (discussed below) can be marked as the threshold of severity level “normal” (medium). Based on the anomalies identified by the anomaly boundary, the threshold that separates the top-N anomalies from other anomalies can be labeled as the threshold of severity level “critical” (high), where N≥1. Additionally, among the anomalies identified by the anomaly boundary, those anomalies that are just marginally beyond the anomaly boundary can be used to derive the threshold of severity level “low.””) is significantly smaller than a number N of log tokens (para [0079] The number of log tokens is more than the number of severity levels.). It would have been obvious to one or ordinary skill in the art before the effective filing date to modify the anomaly detection models of Du and Zhang with the threshold severity levels of Bai. Doing so would allow for identifying the different categories of severity for the anomaly (Bai para [0092]). Regarding Claim 13, Claim 13 is the system corresponding to the method of claim 4. Claim 13 is substantially similar to claim 4 and is rejected on the same grounds. Conclusion Any inquiry concerning this communication or earlier communications from the examiner should be directed to HENRY K NGUYEN whose telephone number is (571)272-0217. The examiner can normally be reached Mon - Fri 9:00am-5:00pm. 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, Li B Zhen can be reached at 5712723768. 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. /HENRY NGUYEN/Primary Examiner, Art Unit 2121
Read full office action

Prosecution Timeline

Jan 18, 2024
Application Filed
Feb 15, 2024
Response after Non-Final Action
Oct 20, 2026
Non-Final Rejection mailed — §101, §103, §112 (current)

Precedent Cases

Applications granted by this same examiner with similar technology

Patent 12783434
TRANSFER DATA IN A MEMORY SYSTEM WITH ARTIFICIAL INTELLIGENCE MODE
7y 2m to grant Granted Nov 03, 2026
Patent 12748946
Method, System, and Computer Program Product for Efficient Content-Based Time Series Retrieval
1y 9m to grant Granted Sep 29, 2026
Patent 12718058
HYBRID NEURAL NETWORK ARCHITECTURE WITHIN CASCADING PIPELINES
5y 8m to grant Granted Aug 25, 2026
Patent 12705302
METHOD, ACCELERATOR, AND ELECTRONIC DEVICE WITH TENSOR PROCESSING
5y 9m to grant Granted Aug 11, 2026
Patent 12704841
DEEP REINFORCEMENT LEARNING-BASED TECHNIQUES FOR END TO END ROBOT NAVIGATION
5y 5m to grant Granted Aug 11, 2026
Study what changed to get past this examiner. Based on 5 most recent grants.

Strategy Recommendation AI-generated — please review before filing

Get a prosecution strategy drawn from examiner precedents, rejection analysis, and claim mapping.
Typically takes 5-10 seconds — AI-generated, attorney review required before filing

Prosecution Projections

1-2
Expected OA Rounds
60%
Grant Probability
86%
With Interview (+26.8%)
4y 5m (~1y 8m remaining)
Median Time to Grant
Low
PTA Risk
Based on 168 resolved cases by this examiner. Grant probability derived from career allowance rate.

Sign in with your work email

Enter your email to receive a magic link. No password needed.

Personal email addresses (Gmail, Yahoo, etc.) are not accepted.

Free tier: 3 strategy analyses per month