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 11/27/2023 .
Claim Objections
The following claim(s) are objected to for formality issues:
Claim 18-20 should be referencing the ion implanter of claim 17, as claim 10 recites a medium. These claims are interpreted as dependent on claim 17 for prior art examination.
Appropriate correction(s) are required.
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, non-transitory computer-readable media, and hardware devices 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 generating predictions for a maintenance recovery time for an ion implanter based on various parameters. This is a mental process of generating inference from data, akin to what may be performed by a semiconductor technician or engineer. In particular:
For claim 1: A method, comprising:
receiving setting parameters for an ion implanter, the setting parameters comprising a set of control parameters corresponding to a set of process parameters for the ion implanter (receiving data for inference can be performed in the mind);
predicting a preventative maintenance (PM) recovery time for a PM recovery phase of the ion implanter based on the setting parameters, the PM recovery time representing a time interval between a start time of the PM recovery phase and an end time of the PM recovery phase, using a machine learning model (generating inferences of time, e.g., based on experience and judgment, may be performed in the mind via a mental model, e.g., mental heuristics); and
presenting the recovery time (presenting results may be performed in the mind) on a graphical user interface (GUI) of an electronic device.
Additional elements are underlined here and analyzed below.
For claim 2: wherein the machine learning model is a variance model comprising an artificial neural network (ANN), where layers of the ANN are trained using output from a stress model (Considering and deriving inferences from a stress model may be performed in the mind).
For claim 3: The method of claim 1, wherein the machine learning model is a control model comprising an artificial neural network (ANN) trained using a first set of training data and re-trained as a variance model using a second set of training data, the first set of training data comprising setting parameters and the second set of training data comprising PM recovery data (Considering and deriving inferences from setting a parameters and recovery data, including updating and retraining a model, may be performed in the mind, e.g., updating mental heuristics for inference).
For claim 4: The method of claim 1, wherein the machine learning model is an artificial neural network (ANN) comprising an input layer, an output layer, and multiple hidden layers, the ANN trained by locking the multiple hidden layers and re-training the input layer and the output layer using PM recovery data, calibration data, or stress data (The leaning and updating of mental models based on consideration of contextual data is a mental process).
For claim 5: The method of claim 1, comprising predicting a start time for a next PM recovery phase of the ion implanter using the machine learning model (Generating inferences from a mental model of the next phase may be performed in the mind).
For claim 6: The method of claim 1, comprising:
predicting the set of process parameters for the ion implanter from the set of control parameters using the machine learning model, wherein the machine learning model is a variance model adapted from a control model using transfer learning (Generating predictions for control parameter may be performed in the mind via judgment, experience; furthermore, building upon and adjust mental heuristics by starting from a transferring from an original model may be performed in the mind);
determining a statistical process control (SPC) limit delta between the predicted process parameters and actual process parameters measured for the ion implanter (Determining significant differences and ranges when dealing with unexpected results may be performed in the mind );
comparing the SPC limit delta to a defined threshold value to obtain a comparison result (comparing results to generate a judgment may be performed in the mind); and
determining the end time of the PM recovery phase based on comparison result (drawing inferences based on earlier results (e.g., via reflection) may be performed in the mind).
For claim 7: The method of claim 1, wherein the control parameter corresponds to a hardware or software setting that controls a configuration or operation of a component of the ion implanter, the at least one control parameter comprising a charge parameter, an energy parameter, an acceleration or deceleration parameter, a dopant and flow parameter, a diluent and flow parameter, a source parameter, an analyzer parameter, a corrector parameter, a suppression parameter, a focus parameter, a scan parameter, a quadrupole lens current parameter, or a post-acceleration voltage parameter (Considering these various parameters during inference may be performed in the mind).
For claim 8: The method of claim 1, wherein the process parameter corresponds to a metric associated with a beam property for an ion beam generated by the ion implanter, the at least one process parameter comprising a beam height parameter, a beam width parameter, full half height maximum (FHHM) parameter, a vertical within device angle (VWIDA) parameter, a VWIDA mean (VWIDAM) parameter, a horizontal within device angle (HWIDA) parameter, a HWIDA mean (HWIDAM) parameter, a standard deviation of VWIDA (VWIDAS) parameter, a standard deviation of HWIDA mean (HWIDAS) parameter, a vertical intensity (VI) parameter, a spotscore parameter, an energy parameter, a region of interest (ROI) current parameter, or a uniformity parameter (considering these various parameters may be performed in the mind).
For claim 9: The method of claim 1, comprising generating instructions to indicate the ion implanter has reached an end time of the PM recovery phase and is ready to enter an operational phase to generate an ion beam for implanting ions in a semiconductor wafer (Generating inference instructions in order to provide indication of a end time or an operation time may be performed in the mind).
Claims 10-20 recite computer readable medium and devices that correspond to the above methods and are hence analyzed in the same way.
STEP 2A PRONG 2: The claims do not integrate the exception into a practical application:
For claim 1, the additional elements comprise displaying of results on a on a GUI of a computing device and the use of a machine learning model to generate the inference. However, this constitutes mere instructions to implement the abstract idea on a general purpose computer and in a machine learning environment and hence do not comprise an integration into a practical application.
For claims 2-3, 5-6, the additional elements comprise the use of a machine learning model comprising a artificial neural network with trainable layers. However, this constitutes mere instructions to implement the abstract idea in machine learning environment using a ANN and does not meaningfully limit the practice of the abstract idea and hence do not comprise an integration into a practical application.
Claim 4 additionally recites a machine learning as an artificial neural network (ANN) comprising an input layer, an output layer, and multiple hidden layers, the training occurring by locking the multiple hidden layers and re-training the input layer and the output layer. However, this constitutes mere instructions to apply the abstract idea in a machine learning environment via transfer learning. It does not impose meaningful limitations on the practice of the abstract idea and hence does not constitute an integration into a practical application.
Claims 10-16 additionally recites: a non-transitory computer-readable storage medium, the computer-readable storage medium including instructions that when executed by a computer, cause the computer to perform a method. However, this constitutes mere instructions to implement the abstract idea on a general purpose computer and in a machine learning environment and hence do not comprise an integration into a practical application.
Claims 17-20 additionally recites: an ion implanter, comprising:
an ion source to generate an ion beam;
at least one beamline component to direct the ion beam towards a substrate;
a processing circuitry; and
a memory coupled to the processing circuitry, the memory storing instructions that, when executed by the processor circuitry, configure the processing circuitry to perform a method.
However, this merely generally links the mentally performable inference technique to a particular field of use, that of ion implantation, and does not impose meaningful limitations on the performance of the mental process. Hence, it does not constitute an integration into a practical application.
The memory and processor elements constitutes mere instructions to implement the abstract idea on a general purpose computer and in a machine learning environment and hence do not comprise an integration into a practical application.
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 comprise displaying of results on a on a GUI of a computing device and the use of a machine learning model to generate the inference. However, the use of computing display and machine learning models is well-understood, routine, and conventional in the field of data inference and hence does not constitute significantly more.
For claims 2-3, 5-6, the additional elements comprise the use of a machine learning model comprising a artificial neural network with trainable layers. However, the use of artificial neural nets with layers as a machine learning model is well-understood, routine, and conventional in the field of data inference and hence does not constitute significantly more.
Claim 4 additionally recites a machine learning as an artificial neural network (ANN) comprising an input layer, an output layer, and multiple hidden layers, the training occurring by locking the multiple hidden layers and re-training the input layer and the output layer. However, this the use of ANN’s and transfer learning is well-understood, routine, and conventional in the field of data inference and hence does not constitute significantly more.
Claims 10-16 additionally recites: a non-transitory computer-readable storage medium, the computer-readable storage medium including instructions that when executed by a computer, cause the computer to perform a method. However, the use of computing display and machine learning models is well-understood, routine, and conventional in the field of data inference and hence does not constitute significantly more.
Claims 17-20 additionally recites: an ion implanter, comprising:
an ion source to generate an ion beam;
at least one beamline component to direct the ion beam towards a substrate;
a processing circuitry; and
a memory coupled to the processing circuitry, the memory storing instructions that, when executed by the processor circuitry, configure the processing circuitry to perform a method.
However, the use of ion implanters is understood, routine, and conventional in the field of semiconductor manufacturing and hence does not constitute significantly more.
Further, the use of computing processor and memory components is well-understood, routine, and conventional in the field of data inference and hence does not constitute significantly more.
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, 5, 7-10, 14, 16-18, 20 are rejected under 35 U.S.C. 103 as being unpatentable over Kitsunai (US 20040147131 A1) in view of Takemura (US 11462385 B2).
For claim 1, Kitsunai discloses: a method, comprising:
receiving setting parameters, the setting parameters comprising a set of control parameters corresponding to a set of process parameters (figs.6-7, 0047: setting parameters, such as a current product and processing conditions are obtained and stored in the database and associated with plasma status, see fig.1:8-9, 0032, various setting parameters corresponding to machine control parameters (see 0032) are gathered and stored in the database for processing, the parameters corresponding the various process parameters including plasma status);
predicting a preventative maintenance (PM) recovery time for a PM recovery phase (fig.6, 0045, 0048) based on the setting parameters (0047-48: PM recovery time is calculated based on historical data including product, conditions, etc. in order to predict recovery time based on threshold, see 0050), the PM recovery time representing a time interval between a start time of the PM recovery phase and an end time of the PM recovery phase (fig.6, 0045, 0048: interval between start time and end time when production may occur), using a machine learning model (fig.6:3, 0048-50: generating computer predictions based on historical conditions including product type, conditions, plasma status, hence, machine learning); and
presenting the recovery time on a graphical user interface (GUI) of an electronic device (fig.7, 0050, 0024).
Kitsunai does not disclose: wherein the setting parameters and process parameters and recover phase are for an ion implanter.
Takemura discloses: wherein the setting parameters and process parameters and recovery phase are for an ion implanter (fig.3, fig.6, c.9¶2-3, ¶7 shows an overview of applying machine learning models to process data for prediction tuning recovery phase for an ion implanter based on various setting and process parameters, see fig.4: recipe settings, basic initial and adjusted process parameter settings for ion implanter).
It would have been obvious before the effective filing date to a person of ordinary skill in the art to modify the method of Kitsunai by incorporating the ion implanter technique of Takemura. Both concern the art of machine learning for semiconductor manufacturing tuning and recovery, and the incorporation would have, according to Takemura, improve ion beam setup time and quality during maintenance (c.1: Description of related art, c.10¶3)
For claim 5, Kitsunai modified by Takemura discloses the method of claim 1, as described above. Kitsunai further discloses: comprising predicting a start time for a next PM recovery phase of the ion implanter using the machine learning model (0044: predicting next interval for maintenance, hence, next start time).
For claim 7, Kitsunai modified by Takemura discloses the method of claim 1, as described above. Takemura further discloses: wherein the control parameter corresponds to a hardware or software setting that controls a configuration or operation of a component of the ion implanter (c.6:15-55: various basic control parameters for ion implanter modules), the at least one control parameter comprising a charge parameter (c.6:30-31, 50: current, voltage), an energy parameter (c.6:30-31: current; c:6:49: flux density), an acceleration or deceleration parameter, a dopant and flow parameter, a diluent and flow parameter (c.6:28-30), a source parameter (c.6:21-28), an analyzer parameter, a corrector parameter, a suppression parameter, a focus parameter (c.6:38-41), a scan parameter, a quadrupole lens current parameter, or a post-acceleration voltage parameter (c.6:50).
For claim 8, Kitsunai modified by Takemura discloses the method of claim 1, as described above. Kitsunai further discloses: wherein the process parameter corresponds to a metric associated with a beam property for an ion beam generated by the ion implanter (fig.2:84, c.7¶2: monitored beam parameters), the at least one process parameter comprising a beam height parameter, a beam width parameter, full half height maximum (FHHM) parameter, a vertical within device angle (VWIDA) parameter, a VWIDA mean (VWIDAM) parameter, a horizontal within device angle (HWIDA) parameter, a HWIDA mean (HWIDAM) parameter, a standard deviation of VWIDA (VWIDAS) parameter, a standard deviation of HWIDA mean (HWIDAS) parameter, a vertical intensity (VI) parameter, a spotscore parameter, an energy parameter (c.7:45-56: filament current and voltage, ), a region of interest (ROI) current parameter, or a uniformity parameter.
For claim 9, Kitsunai discloses the method of claim 1, as described above. Kitsunai further discloses: comprising generating instructions to indicate the ion implanter has reached an end time of the PM recovery phase and is ready to enter an operational phase to generate an ion beam for implanting ions in a semiconductor wafer (Kitsunai fig.7, 0024, 0050; Takemura fig.1).
Claim 10 recites a non-transitory computer-readable medium analogous to the method of claim 1 and is hence rejected for the same reasons. Further, Takemura discloses: a non-transitory computer-readable storage medium, the computer-readable storage medium including instructions that when executed by a computer, cause the computer to perform the method (c.13:35-55).
Claim 17 recites an ion implanter analogous to the method of claim 1 and is hence rejected for the same reasons. Further, Takemura discloses: an ion implanter (fig.1, c.4:55-c.5:43), comprising: an ion source to generate an ion beam (fig.1:2); at least one beamline component to direct the ion beam towards a substrate (fig.1:4); a processing circuitry; and a memory coupled to the processing circuitry, the memory storing instructions that, when executed by the processor circuitry, configure the processing circuitry to perform the method (c.13:35-55).
Claims 14, 16, 18, 20 recite computer media and devices corresponding to the methods 5, 9 above and hence are rejected for the same reasons.
Claim(s) 2-3, 11-12 are rejected under 35 U.S.C. 103 as being unpatentable over Kitsunai (US 20040147131 A1) in view of Takemura (US 11462385 B2) in view of Raj (US 20210334695 A1).
For claim 2, Kitsunai modified by Takemura discloses the method of claim 1, as described above. Kitsunai modified by Takemura further discloses: where the machine learning model is trained using output from a stress model (Takemura c.3¶2-3, c.7¶4-5, c.9:¶2-3, c.13¶3 gives overview of machine learning training including objective variables / explanatory variables and input variables and state parameter, these variables including various indicators of stress including beam angle, filament diameter, filament current, voltage, etc.).
Kitsunai modified by Takemura does not disclose: wherein the machine learning model is a variance model comprising an artificial neural network (ANN), where layers of the ANN are trained.
Nizam discloses: wherein the machine learning model is a variance model comprising an artificial neural network (ANN), where layers of the ANN are trained (p.2: discloses training multiple layers of a neural network including freezing and unfreezing layers to achieve desired training efficiency / accuracy tradeoff based on task similarity considerations).
It would have been obvious before the effective filing date to a person of ordinary skill in the art to modify the method of Kitsunai modified by Takemura by incorporating transfer learning layer freezing technique of Nizam. Both concern the art of machine learning, and the incorporation would have, according to Nizam, achieve desired levels of training efficiency and task specificity based on task similarity (p.2).
For claim 3, Kitsunai modified by Takemura discloses the method of claim 1, as described above. Kitsunai further discloses: wherein the machine learning model is trained using a first set of training data and retrained using a second set of training data, the first set of training data comprising setting parameters (Takemura fig.4, figs.6-7 gives overview of machine learning training including objective variables / explanatory variables and input variables and state parameter, these variables including setting parameters data such as recipe data, initial parameter data, etc.) and the second set of training data comprising PM recovery data (ibid: training data including parameter data such has objective data, e.g., recovery time, see c.3¶2-3, c.7¶4-5, c.9:¶2-3, c.13¶3; the training being iterative, hence, retraining via later data).
Kitsunai modified by Takemura does not disclose: wherein the machine learning model is a control model comprising an artificial neural network (ANN); wherein the retraining is as a variance model using a second set of training data.
Nizam discloses: wherein the machine learning model is a control model comprising an artificial neural network (ANN); wherein the retraining is as a variance model using a second set of training data (p.2: discloses training multiple layers of a neural network including freezing and unfreezing layers to achieve desired training efficiency / accuracy tradeoff based on task similarity considerations, hence, initial and variance model using second training data with some layers frozen).
It would have been obvious before the effective filing date to a person of ordinary skill in the art to modify the method of Kitsunai modified by Takemura by incorporating transfer learning layer freezing technique of Nizam. Both concern the art of machine learning, and the incorporation would have, according to Nizam, achieve desired levels of training efficiency and task specificity based on task similarity (p.2).
Claims 11-12 recite computer media and devices corresponding to the above methods 2-3 and are hence rejected for the same reasons.
Claim(s) 4, 13 are rejected under 35 U.S.C. 103 as being unpatentable over Kitsunai (US 20040147131 A1) in view of Takemura (US 11462385 B2) in view of SeucheAchat ("Finding out ideal layers to freeze and train during transfer learning", published 2020).
For claim 4, Kitsunai modified by Takemura discloses the method of claim 1, as described above. Kitsunai further discloses: wherein the machine learning model is re-trained using PM recovery data, calibration data, or stress data (Takemura fig.4, figs.6-7 gives overview of machine learning training including objective variables / explanatory variables and input variables and state parameter, these variables including calibration data such as recipe data, initial parameter data, etc. and objective data such as PM recovery time, etc.., stress data including filament diameter, see c.3¶2-3, c.7¶4-5, c.9:¶2-3, c.13¶3).
Kitsunai modified by Takemura does not disclose: wherein the machine learning model is an artificial neural network (ANN) comprising an input layer, an output layer, and multiple hidden layers, the ANN trained by locking the multiple hidden layers and retraining the input layer and output layer.
SeucheAchat discloses: wherein the machine learning model is an artificial neural network (ANN) comprising an input layer, an output layer, and multiple hidden layers, the ANN trained by locking the multiple hidden layers and retraining the input layer and output layer (p.2 discloses various strategies for training including freezing and unfreezing a earlier convolutional layers while retraining later fully connected layers, including unfreezing lower layers).
It would have been obvious before the effective filing date to a person of ordinary skill in the art to modify the method of Kitsunai modified by Takemura by incorporating transfer learning layer freezing technique of SeucheAchat. Both concern the art of machine learning, and the incorporation would have, according to SeucheAchat, achieve performance and accuracy tradeoffs (p.2).
Claim(s) 13 recite computer media and devices corresponding to the above methods and are hence rejected for the same reasons.
Claim(s) 6, 15, 19 are rejected under 35 U.S.C. 103 as being unpatentable over Kitsunai (US 20040147131 A1) in view of Takemura (US 11462385 B2) in view of Raj (US 20210334695 A1) in view of Nizam ("What is freezing / unfreezing a layer in neural networks", published 6/6/2020).
For claim 6, Kitsunai modified by Takemura discloses the method of claim 1, as described above. Kitsunai modified by Takemura further discloses: predicting the set of process parameters for the ion implanter from the set of control parameters using the machine learning model (Kitsunai fig.5-6 process parameters such as CD is predicted from process, with Takemura fig.1 disclosing application to ion implanters, see fig.4, fig.6 disclosing machine learning model, c.3¶2-3, c.13¶3, disclosing various predicted process parameter objective variables).
Kitsunai modified by Takemura does not disclose: determining a statistical process control (SPC) limit delta between the predicted process parameters and actual process parameters measured for the ion implanter;
comparing the SPC limit delta to a defined threshold value to obtain a comparison result; and
determining the end time of the PM recovery phase based on comparison result.
Raj discloses: determining a limit delta between the predicted parameters and actual parameters measured (fig.3:308, 314-318, 0054-56: determining an accuracy of the model, hence, determining an accuracy score based on comparing predicted output parameters and actual output parameters, in order to determine a delta change for a limiting threshold comparison; combination with Kitsunai modified by Takemura yielding application to statistical process control values corresponding to statistical predictions of process parameters);
comparing the limit delta to a defined threshold value to obtain a comparison result (0056: obtaining comparison result); and
determining an inference result based on comparison result (ibid: performing model retraining based on the comparison result, hence, determining inference based on a new or old model based on the comparison result indication; combination with Kitsunai modified by Takemura yielding application to PM recovery phase data prediction).
It would have been obvious before the effective filing date to a person of ordinary skill in the art to modify the method of Kitsunai modified by Takemura by incorporating the retraining thresholding technique of Raj. Both concern the art of machine learning, and the incorporation would have, according to Raj, detect and correct data drift (0007).
Kitsunai modified by Takemura modified by Raj does not disclose: wherein the machine learning model is a variance model adapted from a control model using transfer learning.
Nizam discloses: wherein the machine learning model is a variance model adapted from a control model using transfer learning (p.2: discloses training multiple layers of a neural network including freezing and unfreezing multiple middle layers to achieve desired training efficiency / accuracy tradeoff based on task similarity considerations, hence, initial control and adapted variance model trained via transfer learning).
It would have been obvious before the effective filing date to a person of ordinary skill in the art to modify the method of Kitsunai modified by Takemura by incorporating transfer learning layer freezing technique of Nizam. Both concern the art of machine learning, and the incorporation would have, according to Nizam, achieve desired levels of training efficiency and task specificity based on task similarity (p.2).
Claims 15, 19 recite computer media and devices corresponding to the above methods and are hence rejected for the same reasons.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Panda (US 20220214662 A1) discloses machine learning models for predicting seasoning, see fig.6.
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).
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).
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).
/LIANG LI/
Primary examiner AU 2143
Prosecution Insights