DETAILED ACTION
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA . This action is responsive to pending claims 1-20 filed 2/12/2026.
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.
Claim(s) 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The 35 U.S.C. 101 subject matter eligibility analysis first asks whether the claim is directed to one of the four statutory categories (Step 1). It next asks whether the claim is directed to an abstract idea (Step 2A), via Prong 1, whether an abstract idea (e.g., mathematical concept, mental process, certain methods of organizing human activity) is recited, and Prong 2, whether it is integrated into a practical application. It finally asks whether the claim as a whole includes additional elements that amount to significantly more than the judicial exception (Step 2B). See MPEP 2106.
STEP 1: The claims falls within one of the four statutory categories:
All claims are directed to methods, hardware systems, and non-transitory computer-readable media and hence fall within one of the four statutory categories.
STEP 2A PRONG 1: The claims recite a judicial exception:
The claims are directed to sequence or stream or time-series data analysis via recognizing information contexts from the patterns and neighboring patterns in the stream, and determining pairwise causal relationships based on the occurrence of information contexts. As such, it is directed to a mental process, as the recognizing of patterns from data streams and the reasoning from such patterns to form causal relations may be performed in the mind.
The claims additionally recite the performance of these processes on a DPU, separate from a CPU, in order to reduce CPU overhead. However, at present this broadly claims any performance on a DPU, as the negative limitations (“CPU external to the storage device”, “without transferring … to the CPU”, etc.) simply recite the independent performance on a DPU, with a decoupled CPU resource exiting somewhere in the system but unrelated to the operation of the DPU. Hence, for the reasons given below, they do not constitute significantly more.
The additional elements (underlined below) recite implementation on DPUs and details involving storage. However, for the reasons given below, they do not constitute an integration into a practical application (2a-2) nor significantly more (2b). In particular:
For claim 1, discloses A method of offloading knowledge base creation into a storage to reduce central processing unit (CPU) overhead, comprising (The creation of knowledge bases by using observation and judgment to infer causal principles from patterns is a mental process; furthermore, as such rules may be remembered in the mind or with pen and paper, offloading into a storage space is a mental process), comprising:
identifying a sequence of patterns in a data stream based on a time dimension of the data stream by data processing units (DPUs) located in the storage device without an initiation from a CPU external to the storage device, wherein the time dimension reflects a chronological ordering of data items in the data stream (As described above, identifying series patterns time-series data is a mental process.);
recognizing, by the DPUs, a plurality of information contexts corresponding to the sequence of patterns based on analyzing neighboring patterns of any particular pattern in the sequence of patterns (recognizing recurring contexts or neighborhoods in time-series data may be performed in the mind); and
determining, by the DPUs, causal relations among the sequence of patterns based on detecting repetitions of any pair of information contexts among the plurality of information contexts, wherein the causal relations comprise a plurality of reason-consequence pairs; (determining or deriving causal relations from the observed data context repetition may be performed in the mind);
generating, by the DPUs, predictions of future states of the data stream based on at least a portion of the causal relations without transferring the data stream to the CPU (generating predictions based on inferred causal relations, such as based on a mental heuristic, is a mental process).
For claim 2: The method of claim 1, further comprising:
detecting, by at least one of the DPUs, all instances of a particular information context's relations with remaining information contexts among the plurality of information contexts (detecting or observation of particular information contexts may be performed in the mind; see above for analysis of DPU implementation); and
detecting, by the at least one of the DPUs, whether any detected relations are reproducible (detecting properties or repetitions of patterns may be performed in the mind).
For claim 3: The method of claim 2, further comprising:
generalizing reproducible relations to the causal relations (generalizing patterns to relations or rules is a mental process); and
storing the causal relations into a knowledge base in the storage device (remembering rules is a mental process).
For claim 4: The method of claim 1, further comprising:
creating a time map of causality by connecting a subset of the plurality of reason- consequence pairs into a sequence based on the time dimension of the data stream (synthesizing causal relations to generate a map or network based on observed data is a mental process).
For claim 5: The method of claim 4, further comprising:
detecting whether the sequence of reason-consequence pairs is reproducible based at least in part on a predetermined similarity threshold (detecting reproducibility of observed pairs based on some similarity observation is a mental process); and
generalizing a reproducible sequence of reason-consequence pairs and storing the generalized sequence of reason-consequence pairs into a knowledge base in the storage device (generalizing and storing (e.g., remembering, writing down) such generalizations of observed pair patterns is a mental process).
For claim 6: The method of claim 1, wherein the generating the predictions of future states of the data stream based on at least the portion of the causal relations comprises generating the predictions indicative of the future states of the data stream based on a knowledge base created in the storage device (Generating predictions from remembered data is a mental process, akin to reasoning from experience).
For claim 7: The method of claim 6, further comprising:
determining whether the predictions are correct based on comparing the predictions with new future states associated with the data stream (Comparing and checking for consistency is a mental process); and
storing at least a subset of the predictions to the knowledge base in response to determining that the at least a subset of the predictions are correct (Storing or remembering patterns and rules in response to such comparisons is a mental process).
For claim 8. The method of claim 7, further comprising:
generating hypotheses associated with the data stream based on the knowledge base (Generating hypotheses based on observed patterns is a mental process).
The remaining claims recite systems and computer-readable media corresponding to the above claims and hence are likewise rejected.
STEP 2A PRONG 2: The claims do not integrate the exception into a practical application:
For claim 1, the additional elements recite implementation on a hardware system comprising a DPU, in order to reduce DPU overhead, the identifying occurring without initiation from a CPU external to the storage device, the generating occurring without transferring the data stream to the CPU. However, as these elements merely speak of the coexistence of a decoupled DPU and CPU, i.e., the negative limitations only recite that the CPU does not perform certain steps, in effect, saying that the process occurs independently on the DPU. However, this constitutes mere instructions to implement the abstract idea on a DPU and hence do not comprise an integration into a practical application.
For claim 9 recites: a system, comprising: at least one processor; and at least one memory comprising computer-readable instructions that upon execution by the at least one processor cause the computing device to perform operations. However, these constitute mere instructions to implement an abstract idea on a computer and hence does not constitute an integration into a practical application (2a-2).
Claim 15 recites: a non-transitory computer-readable storage medium, storing computer-readable instructions that upon execution by a processor cause the processor to implement operations. However, this constitute mere instructions to implement an abstract idea on a computer and hence does not constitute an integration into a practical application (2a-2).
STEP 2B: The claim as a whole do not include additional elements that amount to significantly more than the abstract idea:
For claim 1, the additional elements recite implementation on a hardware system comprising a DPU, in order to reduce DPU overhead, the identifying occurring without initiation from a CPU external to the storage device, the generating occurring without transferring the data stream to the CPU. However, as these elements merely speak of the coexistence of a decoupled DPU and CPU, i.e., the negative limitations only recite that the CPU does not perform certain steps, in effect, saying that the process occurs independently on the DPU. However, DPUs are well-understood, routine, and conventional and hence do not comprise an integration into a practical application.
For claim 9 recites: a system, comprising: at least one processor; and at least one memory comprising computer-readable instructions that upon execution by the at least one processor cause the computing device to perform operations. However, the use of computers is well-understood, routine and conventional (WURC) and hence does not constitute significantly more (2b).
Claim 15 recites: a non-transitory computer-readable storage medium, storing computer-readable instructions that upon execution by a processor cause the processor to implement operations. However, the use of computers is well-understood, routine and conventional (WURC) and hence does not constitute significantly more (2b).
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.
Claim(s) 1-6, 9-12, 15-18 are rejected under 35 U.S.C. 103 as being unpatentable over Chattopadhyay (US 10817794 B1) in view of Niu (US 20180121120 A1).
For claim 1, Chattopadhyay discloses: a method of offloading knowledge base creation, comprising (fig.2 gives overview of a method for creating model-based knowledge base creation):
identifying a sequence of patterns in a data stream based on a time dimension of the data stream (fig.10 ¶1: acquiring input data streams, i.e., patterns, to derive causal dependance, hence, temporally; see also p.10 ¶3: temporal series; see also examples in seismic prediction (p.16 last ¶), stock market, neural connectivity (p.17 ¶1-3), hence, temporal signals), wherein the time dimension reflects a chronological ordering of data items in the data stream (ibid);
recognizing a plurality of information contexts corresponding to the sequence of patterns based on analyzing neighboring patterns of any particular pattern in the sequence of patterns (fig.3A, c.11:last ¶ discloses the construction of state transition models PFSA from input streams, see also c.12¶4-10: determination of context sequences, hence, recognizing information contexts based on analyzing neighboring a patterns of a particular context pattern in the sequence of patterns); and
determining causal relations among the sequence of patterns based on detecting repetitions of any pair of information contexts among the plurality of information contexts, wherein the causal relations comprise a plurality of reason-consequence pairs (c.12¶4-10: a PFSA is constructed, this PFSA being a diagram of causal relationships between patten sequences and being constructed based on observations of patterns in order to construct transition probabilities between pairs of contexts, the causal relations being a plurality of causal or reason-consequence pairs); and
generating predictions of future states of the data stream based on at least a portion of the causal relations (fig.2, c.10 ¶1: generating predictions based on inferred PFSAs; see also c.13:25-35).
Chattopadhyay does not disclose: wherein the offloading is into a storagewherein the identifying is by data processing units (DPUs) located in the storage device without an initiation from a CPU external to the storage device; wherein the recognizing, determining, generating are by the DPUs; wherein the generating takes place without transferring the data stream to the CPU.
Niu discloses: wherein the offloading is into a storage(fig.6, 0082-86 contemplates ad hoc processing of jobs via DPUs, such an ad hoc construction hence reducing CPU overhead, such as in a hybrid network (0084-85)); wherein the identifying is by data processing units (DPUs) located in the storage device without an initiation from a CPU external to the storage device (ibid, particularly 0084-85: ad hoc control of jobs without CPU initiation, hence, combination with the tasks of Chattopadhyay yielding performing of identifying in such a manner); wherein the recognizing, determining, generating are by the DPUs (ibid); wherein the generating takes place without transferring the data stream to the CPU (ibid).
It would have been obvious before the effective filing date to a person of ordinary skill in the art to modify the method of Chattopadhyay by incorporating the DPU cluster hardware of Niu. Both concern the art of machine learning, and the incorporation would have, according to Niu, provide greater processing capacity via reconfigurable DPU pools (0085).
For claim 2, Chattopadhyay modified by Niu discloses the method of claim 1, as described above. Chattopadhyay modified by Niu further discloses: detecting, by at least one of the DPUs (Niu fig.6), all instances of a particular information context's relations with remaining information contexts among the plurality of information contexts (Chattopadhyay c.12¶4-10 contemplates construction of a PFSA to model the entire ergodic system); and detecting, by the at least one of the DPUs, whether any detected relations are reproducible (ibid: contemplates determining causation based on repeated observations, with fig.2 contemplating prediction, c.16 ¶5 (“Model Validation”) contemplating validation of reproducibility).
For claim 3, Chattopadhyay modified by Niu discloses the method of claim 2, as described above. Chattopadhyay modified by Niu further discloses: generalizing reproducible relations to the causal relations (Chattopadhyay c.12¶4-10: generalizing reproducible information probabilistically to PFSA causal relations); and storing the causal relations into a knowledge base in the storage device (fig.2: models are stored for prediction, such has via hardware of fig.1A-B, with Niu fig.6 disclosing the use of DPU storage devices).
For claim 4, Chattopadhyay modified by Niu discloses the method of claim 1, as described above. Chattopadhyay modified by Niu further discloses: creating a time map of causality by connecting a subset of the plurality of reason-consequence pairs into a sequence based on the time dimension of the data stream (c.12¶4-10: the PFSAs constitute time maps of causality of directional sequences).
For claim 5, Chattopadhyay modified by Niu discloses the method of claim 4, as described above. Chattopadhyay modified by Niu further discloses: detecting whether the sequence of reason-consequence pairs is reproducible based at least in part on a predetermined similarity threshold (Chattopadhyay c.12¶4-10 discloses construction of PFSA which is a directed graph comprising linked reason-consequence pairs, each corresponding to a particular string or context string, the algorithm determining a set of links between states based on a clustering distance threshold, see c.12 step (3), hence, the reproducibility of the various reason-consequence pair states, whether they are in fact in the PFSA, being based on a cluster similarity threshold); and generalizing a reproducible sequence of reason-consequence pairs and storing the generalized sequence of reason-consequence pairs into a knowledge base (fig.2: creating model, see also c.12 step (5)) in the storage device (Chattopadhyay fig.1B, Niu fig.1-2).
For claim 6, Chattopadhyay modified by Niu discloses the method of claim 1, as described above. Chattopadhyay modified by Niu further discloses: wherein the generating the predictions of future states of the data stream based on at least the portion of the causal relations comprises generating the predictions indicative of the future states of the data stream based on a knowledge base created in the storage device (Chattopadhyay fig.2, c.10¶1, c.13:25-35): predictions are generate based on the PFSA causality model, with Niu figs.1-2 disclosing storage device).
Claims 9-12, 15-18 recite systems and computer media corresponding to the above methods and are hence likewise rejected. Furthermore, Chattopadhyay discloses:
for claim 9: a system, comprising:
at least one processor (fig.1B:101); and
at least one memory comprising computer-readable instructions that upon execution by the at least one processor cause the computing device to perform operations (fig.1B:102); for claim 15: a non-transitory computer-readable storage medium, storing computer-readable instructions that upon execution by a processor cause the processor to implement operations (fig.1B:101-102, c.9¶4).
Claim(s) 7-8, 13-14, 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Chattopadhyay (US 10817794 B1) in view of Niu (US 20180121120 A1) in view of Sato (US 20180189654 A1).
For claim 7, Chattopadhyay modified by Niu discloses the method of claim 6, as described above. Chattopadhyay modified by Niu does not discloses: determining whether the predictions are correct based on comparing the predictions with new future states associated with the data stream; and
storing at least a subset of the predictions to the knowledge base in response to determining that the at least a subset of the predictions are correct.
Sato discloses: determining whether the predictions are correct based on comparing the predictions with new future states associated with the data stream (fig.3, 0028-31: collecting new data and calculating updated accuracy data of predictors); and
storing at least a subset of the predictions to the knowledge base in response to determining that the at least a subset of the predictions are correct (fig.3:124, 0032; fig.4:103, 0040: predictions corresponding to the correct predictions are ranked and sorted for storage).
It would have been obvious before the effective filing date to a person of ordinary skill in the art to modify the method of Chattopadhyay modified by Niu by incorporating the prediction updating and storage technique of Sato. Both concern the art of predictive machine learning, and the incorporation would have, according to Sato, raise predictive accuracy and robustness via multiple models (0067-68).
For claim 8, Chattopadhyay modified by Niu modified by Sato discloses the method of claim 7, as described above. Chattopadhyay modified by Niu further discloses: generating hypotheses associated with the data stream based on the knowledge base (fig.12, c.10¶1, c.16 last ¶-c.17¶4: generating predictive hypotheses based on knowledge base).
Claims 13-14, 19-20 recite systems and computer media corresponding to the above methods and are hence likewise rejected. Furthermore, Chattopadhyay discloses:
Response to Arguments
In the remarks, Applicant argued:
For the 101 rejection:
1) The 101 is not evoked because the steps could not be performed in the mind, or practically, with pen and paper the recited steps, especially as they are performed by a DPU with respect to a CPU.
Examiner respectfully disagrees. At present, apart from the hardware implementation, the human mind is able to perform all the steps of identifying patterns, recognizing contextual occurrences, determining relations among patterns, and generating predictions. This corresponds to sequence recognition and prediction which the mind performs in many daily contexts, e.g., in team sports, video games, solving puzzles, listening to music, writing a paper, etc. Hence, Examiner submits that the claims are directed to a mental process.
2) The claims are integrated into a practical application because they are inextricably tied to a technique for reducing CPU overhead.
As pointed out above in the 101 rejection, the hardware implementation being recited comprises a series of negative limitations regarding the CPU (“external to …”, “without initiation from …”, “without transferring …”) and hence speaks essentially to a decoupled DPU-CPU system. As such, the BRI of such an implementation is simply direct implementation on a DPU in a decouple CPU-DPU system, the implementation thereby serving to reduce CPU processing by not involving it. Hence, they comprise mere instructions for implementing a mental process on a DPU, a conventional and well-known processing device, and hence cannot be said to be an integration into a practical application nor significantly more.
For the 103 rejections:
1) Karas does not disclose the clustering being identifying as sequence of patterns based on a time dimension of the data stream.
2) Karas does not disclose a sequence of patterns, nor recognizing a plurality of information contexts based on analyzing neighboring patterns o a particular pattern.
3) Karas does not determine causal relations among the pattern sequences.
4) Barsellotti does not disclose DPUs located in the storage device.
Applicant’s arguments have been fully considered but are moot in view of the newly cited art.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Dubeyko (US 20200151020 A1) discloses data processing architecture based on DPUs.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any extension fee pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to LIANG LI whose telephone number is (303)297-4263. The examiner can normally be reached Mon-Fri 9-12p, 3-11p MT (11-2p, 5-1a ET).
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor Jennifer Welch can be reached on (571)272-7212. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from Patent Center and the Private Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from Patent Center or Private PAIR. Status information for unpublished applications is available through Patent Center or Private PAIR to authorized users only. Should you have questions about access to Patent Center or the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
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. The examiner is available for interviews Mon-Fri 6-11a, 2-7p MT (8-1p, 4-9p ET).
/LIANG LI/
Primary examiner AU 2143
Prosecution Insights