USPTO Examiner LEE TSU CHANG - Art Unit 2128

Recent Applications

Detailed information about the 100 most recent patent applications.

Application NumberTitleFiling DateDisposal DateDispositionTime (months)Office ActionsRestrictionsInterviewAppeal
19319633THRESHOLD-BASED ADAPTIVE ONTOLOGY AND KNOWLEDGE GRAPH MODIFICATION USING GENERATIVE ARTIFICIAL INTELLIGENCESeptember 2025March 2026Allow600NoNo
19265958HYBRID LANGUAGE MODEL AND DETERMINISTIC PROCESSING FOR UNCERTAINTY ANALYSISJuly 2025December 2025Allow510YesNo
19196702ANOMALY DETECTION METHOD FOR MODEL OUTPUTSMay 2025February 2026Allow920YesNo
19062786HALLUCINATION MITIGATION THROUGH MODULAR MODEL ENSEMBLESFebruary 2025February 2026Allow1120YesNo
19052538FAULT DETECTION METHOD FOR REFRIGERATION UNITS BASED ON IMPROVED DEEP LEARNING MODELFebruary 2025August 2025Allow620NoNo
19019273INTEGRATED CIRCUIT TO IMPLEMENT SPIKING NEURAL NETWORKJanuary 2025January 2026Allow1220NoNo
19000716SYSTEMS AND METHODS FOR PARALLEL EXPLORATION OF A HYPERPARAMETER SEARCH SPACEDecember 2024November 2025Allow1130YesNo
19000685SYSTEMS AND METHODS FOR PARALLEL EXPLORATION OF A HYPERPARAMETER SEARCH SPACEDecember 2024July 2025Allow710YesNo
18923064ARTIFICIAL INTELLIGENCE TECHNIQUES UTILIZING A GENERATIVE RELATIONAL NETWORKOctober 2024April 2025Allow620NoNo
18923458METHOD AND SYSTEM FOR LOCAL COMPRESSION OF ARTIFICIAL INTELLIGENCE MODELOctober 2024March 2025Allow510YesNo
18918265DISTRIBUTED NONLINEAR SUPPORT VECTOR MACHINESOctober 2024March 2025Allow510YesNo
18823415RICH MEDIA PRESENTATION OF RECOMMENDATIONS IN GENERATIVE MEDIASeptember 2024September 2025Allow1220NoNo
18823403INFORMATION PROCESSING METHOD AND APPARATUSSeptember 2024July 2025Allow1120NoNo
18798724SYSTEMS AND METHODS FOR INTERFACING WITH A SERVER OVER A NETWORKAugust 2024October 2025Allow1531YesNo
18833603Method, System, and Computer Program Product for Improving Machine Learning ModelsJuly 2024March 2026Allow1930YesNo
18770545Cross-Model Format ComparisonJuly 2024March 2026Allow2030NoYes
18752163NEURAL NETWORK METHOD AND APPARATUSJune 2024December 2025Allow1830YesNo
18739449System, Method, and Computer Program Product for Time-Based Ensemble Learning Using Supervised and Unsupervised Machine Learning ModelsJune 2024December 2025Allow1820YesNo
18737595MESSAGE SUGGESTIONSJune 2024February 2026Allow2020YesNo
18736250SYSTEM AND METHOD FOR DEVICE IDENTIFICATION AND UNIQUENESSJune 2024April 2025Allow1010NoNo
18732052METHODS AND SYSTEMS FOR NEURAL ARCHITECTURE SEARCHJune 2024February 2026Allow2130NoNo
18648707AUTOMATED SYSTEMS FOR MACHINE LEARNING MODEL DEVELOPMENT, ANALYSIS, AND REFINEMENTApril 2024October 2025Allow1820YesNo
18632180HIGH-DENSITY NEUROMORPHIC COMPUTING ELEMENTApril 2024February 2025Allow1010NoNo
18615189Computer-implemented or hardware-implemented method of entity identification, a computer program product and an apparatus for entity identificationMarch 2024October 2025Allow1920NoNo
18431472Method, Device and Equipment for Selecting Key Geological Parameters of a To-Be-Prospected BlockFebruary 2024October 2024Abandon820YesNo
18394205SYSTEMS AND METHODS FOR BLIND MULTIMODAL LEARNINGDecember 2023February 2025Allow1410NoNo
18495902MACHINE LEARNING FOR NUTRIENT QUANTITY ESTIMATION IN SCORE-BASED DIETS AND METHODS OF USE THEREOFOctober 2023April 2025Abandon1810NoNo
18378649METHOD FOR GENERATING PROGRAMMABLE ACTIVATION FUNCTION AND APPARATUS USING THE SAMEOctober 2023March 2025Allow1710NoNo
18461473FLEET AND ASSET MANAGEMENT FOR EDGE COMPUTING OF MACHINE LEARNING AND ARTIFICIAL INTELLIGENCE WORKLOADS DEPLOYED FROM CLOUD TO EDGESeptember 2023June 2024Allow921YesNo
18360129ONLINE PROBABILISTIC INVERSE OPTIMIZATION SYSTEM, ONLINE PROBABILISTIC INVERSE OPTIMIZATION METHOD, AND ONLINE PROBABILISTIC INVERSE OPTIMIZATION PROGRAMJuly 2023March 2026Allow3130NoNo
18226858System, Method, and Computer Program Product for Implementing a Hybrid Deep Neural Network Model to Determine a Market StrategyJuly 2023May 2024Allow1010NoNo
18271231METHOD FOR ANALYZING OUTPUT OF NEURAL NETWORK, AND SYSTEM THEREFORJuly 2023March 2026Allow3200NoNo
18209188TRADE PLATFORM WITH REINFORCEMENT LEARNINGJune 2023August 2024Allow1510NoNo
18207043PROCESSING MACHINE LEARNING ATTRIBUTESJune 2023April 2025Allow2250YesNo
18329418TRANSPOSING NEURAL NETWORK MATRICES IN HARDWAREJune 2023September 2024Allow1510NoNo
18324453DETERMINING JOURNALIST RISK OF A DATASET USING POPULATION EQUIVALENCE CLASS DISTRIBUTION ESTIMATIONMay 2023April 2025Abandon2330YesNo
18310658AUTOMATED SYSTEMS FOR MACHINE LEARNING MODEL DEVELOPMENT, ANALYSIS, AND REFINEMENTMay 2023October 2023Allow520YesNo
18309106PARTITIONING SENSOR BASED DATA TO GENERATE DRIVING PATTERN MAPApril 2023May 2024Allow1210NoNo
18308784METHOD AND SYSTEM FOR DETERMINING CONVERTER TAPPING QUANTITYApril 2023January 2024Allow920NoNo
18303134NEURAL NETWORK FOR PROCESSING GRAPH DATAApril 2023May 2024Allow1310NoNo
18133800SYSTEM AND METHOD FOR DEVICE IDENTIFICATION AND UNIQUENESSApril 2023March 2024Allow1110NoNo
18175117SUPPORT MODEL AND ORCHESTRATION PROCEDURE FOR MANUAL LABELING PROCESSESFebruary 2023March 2026Allow3610NoNo
18174343LEARNING APPARATUS AND METHODFebruary 2023March 2026Allow3610YesNo
18173005LEARNING APPARATUS AND METHODFebruary 2023March 2026Allow3710YesNo
18112827DATA OPTIMIZATION FOR HIGH BANDWIDTH (HBW) NVM AI INFERENCE SYSTEMFebruary 2023November 2025Allow3200NoNo
18111471HIGH-DENSITY NEUROMORPHIC COMPUTING ELEMENTFebruary 2023January 2024Allow1100YesNo
18019119ARTIFICIAL INTELLIGENCE FEEDBACK METHOD AND ARTIFICIAL INTELLIGENCE FEEDBACK SYSTEMFebruary 2023March 2026Allow3710NoNo
18162204PROVISION OF COMPUTER RESOURCES BASED ON LOCATION HISTORYJanuary 2023April 2024Allow1510NoNo
18103483EVALUATING MACHINE LEARNING MODEL PERFORMANCE BY LEVERAGING SYSTEM FAILURESJanuary 2023July 2023Allow610YesNo
18100455METHOD FOR REDUCING BIAS IN DEEP LEARNING CLASSIFIERS USING ENSEMBLESJanuary 2023January 2026Allow3510NoNo
18154551COMPUTERIZED ASSISTANCE USING ARTIFICIAL INTELLIGENCE KNOWLEDGE BASEJanuary 2023April 2024Allow1510YesNo
18153220REAL-TIME ENSEMBLE EVALUATIONJanuary 2023January 2026Allow3610YesNo
18014428INFERENCE DEVICE, INFERENCE METHOD, AND RECORDING MEDIUMJanuary 2023March 2026Allow3810YesNo
18092256TELEOPERATION FOR TRAINING OF ROBOTS USING MACHINE LEARNINGDecember 2022November 2025Allow3400NoNo
18087290GENERATING HIGH-QUALITY THREAT INTELLIGENCE FROM AGGREGATED THREAT REPORTSDecember 2022January 2026Allow3610NoNo
18081721Controller training based on historical dataDecember 2022March 2024Allow1510YesNo
18080777BINARY NEURAL NETWORK MODEL TRAINING METHOD AND SYSTEM, AND IMAGE PROCESSING METHOD AND SYSTEMDecember 2022February 2026Abandon3811NoNo
18075784LEARNING HYPER-PARAMETER SCALING MODELS FOR UNSUPERVISED ANOMALY DETECTIONDecember 2022January 2026Allow3710YesNo
18061697GRAPH CONVOLUTIONAL NETWORKS WITH MOTIF-BASED ATTENTIONDecember 2022September 2024Allow2220YesNo
18008293Scalable Transfer Learning with Expert ModelsDecember 2022January 2026Allow3710YesNo
17981243SCALABLE SYSTEMS AND METHODS FOR CURATING USER EXPERIENCE TEST RESULTSNovember 2022July 2023Allow920YesNo
17980706MULTIPLATE CONDUCTORNovember 2022July 2025Allow3300NoNo
17981264Automatic Generation of Preferred Views for Personal Content CollectionsNovember 2022July 2025Abandon3220YesYes
17976679LEAN PARSING: A NATURAL LANGUAGE PROCESSING SYSTEM AND METHOD FOR PARSING DOMAIN-SPECIFIC LANGUAGESOctober 2022March 2024Allow1710YesNo
17975198CONTROL METHOD BASED ON ADAPTIVE NEURAL NETWORK MODEL FOR DISSOLVED OXYGEN OF AERATION SYSTEMOctober 2022May 2023Allow720NoNo
17972510EXPLANATORY DROPOUT FOR MACHINE LEARNING MODELSOctober 2022November 2025Allow3610YesNo
18046906COMPUTER-IMPLEMENTED OR HARDWARE-IMPLEMENTED METHOD OF ENTITY IDENTIFICATION, A COMPUTER PROGRAM PRODUCT AND AN APPARATUS FOR ENTITY IDENTIFICATIONOctober 2022April 2023Allow610NoNo
17966323ARCHIVING FILES VIA SUPERDENSE CODINGOctober 2022October 2025Allow3601NoNo
17938686SOLID-STATE DETECTOR CHARACTERIZATION BY MACHINE LEARNING-BASED PHYSICAL MODEL WITH REDUCED DEFECT LEVELSOctober 2022November 2025Allow3711NoNo
17953517METHOD FOR MULTI-TASK-BASED PREDICTING MASSIVEUSER LOADS BASED ON MULTI-CHANNEL CONVOLUTIONAL NEURAL NETWORKSeptember 2022June 2023Abandon910NoNo
17940798MACHINE LEARNING FOR NUTRIENT QUANTITY ESTIMATION IN SCORE-BASED DIETS AND METHODS OF USE THEREOFSeptember 2022June 2023Allow920YesNo
17939351SYSTEMS AND METHODS FOR BLIND MULTIMODAL LEARNINGSeptember 2022August 2023Allow1110NoNo
17821910TRANSFERRING INFORMATION THROUGH KNOWLEDGE GRAPH EMBEDDINGSAugust 2022June 2025Allow3400NoNo
17886499COMPUTER-READABLE RECORDING MEDIUM HAVING STORED THEREIN MACHINE LEARNING PROGRAM, METHOD FOR MACHINE LEARNING, AND INFORMATION PROCESSING APPARATUSAugust 2022March 2026Allow4320YesNo
17878242LEARNING DEVICE, INFORMATION PROCESSING APPARATUS, SUBSTRATE PROCESSING DEVICE, SUBSTRATE PROCESSING SYSTEM, LEARNING METHOD, RECIPE DETERMINATION METHOD AND NON-TRANSITORY COMPUTER-READABLE MEDIUM STORING LEARNING PROGRAMAugust 2022July 2025Allow3501NoNo
17795233NEURAL HASHING FOR SIMILARITY SEARCHJuly 2022June 2023Allow1010NoNo
17814782STICKIFICATION USING ANYWHERE PADDING TO ACCELERATE DATA MANIPULATIONJuly 2022September 2025Allow3800NoNo
17871389Image Processing Model Training Method and ApparatusJuly 2022December 2025Abandon4110NoNo
17864885FINGERPRINTING DATA TO DETECT VARIANCESJuly 2022March 2026Allow4421YesNo
17791967LEARNING PROCESSING DEVICE AND LEARNING PROCESSING METHODJuly 2022March 2026Allow4410NoNo
17807761UNCERTAINTY SCORING FOR NEURAL NETWORKS VIA STOCHASTIC WEIGHT PERTURBATIONSJune 2022February 2026Abandon4420YesNo
17785429SYSTEM, METHOD, AND COMPUTER PROGRAM PRODUCT FOR TIME-BASED ENSEMBLE LEARNING USING SUPERVISED AND UNSUPERVISED MACHINE LEARNING MODELSJune 2022March 2024Allow2120YesNo
17750568METHOD FOR GENERATING PROGRAMMABLE ACTIVATION FUNCTION AND APPARATUS USING THE SAMEMay 2022July 2023Allow1420NoNo
17750072COUPLED NETWORKS FOR PHYSICS-BASED MACHINE LEARNINGMay 2022July 2025Allow3800NoNo
17748891CAUSAL INFERENCE VIA NEUROEVOLUTIONARY SELECTIONMay 2022November 2025Allow4220YesNo
17744601NON-UNIFORM QUANTIZATION OF PRE-TRAINED DEEP NEURAL NETWORKMay 2022March 2023Allow1010NoNo
17769707METHOD AND APPARATUS FOR PRUNING NEURAL NETWORKSApril 2022March 2026Allow4720NoNo
17659015HARMONIC DENSELY CONNECTING METHOD OF BLOCK OF CONVOLUTIONAL NEURAL NETWORK MODEL AND SYSTEM THEREOF, AND NON-TRANSITORY TANGIBLE COMPUTER READABLE RECORDING MEDIUMApril 2022May 2025Allow3700NoNo
17719314INTELLIGENT TOPIC SEGMENTATION WITHIN A COMMUNICATION SESSIONApril 2022January 2025Allow3320YesNo
17716945HIGH PERFORAMANCE MACHINE LEARNING INFERENCE FRAMEWORK FOR EDGE DEVICESApril 2022February 2023Allow1110NoNo
17712876Localized Temporal Model ForecastingApril 2022April 2024Allow2530YesNo
17710663SEMANTIC SCRIPT LANGUAGE PROCESSINGMarch 2022July 2025Allow4020YesNo
17706586ASYNCHRONOUS NEURAL NETWORK TRAININGMarch 2022May 2024Allow2620NoNo
17763739Field Programmable Gate Array Architecture Optimized For Machine Learning ApplicationsMarch 2022January 2026Allow4511YesNo
17703935MULTI-CHANNEL PROTEIN VOXELIZATION TO PREDICT VARIANT PATHOGENICITY USING DEEP CONVOLUTIONAL NEURAL NETWORKSMarch 2022May 2025Allow3810YesNo
17681480Processing Machine Learning AttributesFebruary 2022March 2023Allow1210NoNo
17677556HYPERBOLIC FUNCTIONS FOR MACHINE LEARNING ACCELERATIONFebruary 2022April 2023Allow1310YesNo
17635405METHODS, APPARATUS AND MACHINE-READABLE MEDIA RELATING TO MACHINE-LEARNING IN A COMMUNICATION NETWORKFebruary 2022November 2025Allow4511NoNo
17635210A COMPUTER-IMPLEMENTED OR HARDWARE-IMPLEMENTED METHOD OF ENTITY IDENTIFICATION, A COMPUTER PROGRAM PRODUCT AND AN APPARATUS FOR ENTITY IDENTIFICATIONFebruary 2022July 2022Allow500NoNo
17591161CORRECTING CONTENT GENERATED BY DEEP LEARNINGFebruary 2022October 2024Allow3310YesNo

Appeals Overview

This analysis examines appeal outcomes and the strategic value of filing appeals for examiner LEE, TSU-CHANG.

Patent Trial and Appeal Board (PTAB) Decisions

Total PTAB Decisions
13
Examiner Affirmed
6
(46.2%)
Examiner Reversed
7
(53.8%)
Reversal Percentile
80.6%
Higher than average

What This Means

With a 53.8% reversal rate, the PTAB has reversed the examiner's rejections more often than affirming them. This reversal rate is in the top 25% across the USPTO, indicating that appeals are more successful here than in most other areas.

Strategic Value of Filing an Appeal

Total Appeal Filings
36
Allowed After Appeal Filing
13
(36.1%)
Not Allowed After Appeal Filing
23
(63.9%)
Filing Benefit Percentile
59.2%
Higher than average

Understanding Appeal Filing Strategy

Filing a Notice of Appeal can sometimes lead to allowance even before the appeal is fully briefed or decided by the PTAB. This occurs when the examiner or their supervisor reconsiders the rejection during the mandatory appeal conference (MPEP § 1207.01) after the appeal is filed.

In this dataset, 36.1% of applications that filed an appeal were subsequently allowed. This appeal filing benefit rate is above the USPTO average, suggesting that filing an appeal can be an effective strategy for prompting reconsideration.

Strategic Recommendations

Appeals to PTAB show good success rates. If you have a strong case on the merits, consider fully prosecuting the appeal to a Board decision.

Filing a Notice of Appeal is strategically valuable. The act of filing often prompts favorable reconsideration during the mandatory appeal conference.

Examiner LEE, TSU-CHANG - Prosecution Strategy Guide

Executive Summary

Examiner LEE, TSU-CHANG works in Art Unit 2128 and has examined 228 patent applications in our dataset. With an allowance rate of 81.6%, this examiner has an above-average tendency to allow applications. Applications typically reach final disposition in approximately 44 months.

Allowance Patterns

Examiner LEE, TSU-CHANG's allowance rate of 81.6% places them in the 53% percentile among all USPTO examiners. This examiner has an above-average tendency to allow applications.

Office Action Patterns

On average, applications examined by LEE, TSU-CHANG receive 2.15 office actions before reaching final disposition. This places the examiner in the 58% percentile for office actions issued. This examiner issues a slightly above-average number of office actions.

Prosecution Timeline

The median time to disposition (half-life) for applications examined by LEE, TSU-CHANG is 44 months. This places the examiner in the 14% percentile for prosecution speed. Applications take longer to reach final disposition with this examiner compared to most others.

Interview Effectiveness

Conducting an examiner interview provides a +5.3% benefit to allowance rate for applications examined by LEE, TSU-CHANG. This interview benefit is in the 31% percentile among all examiners. Recommendation: Interviews provide a below-average benefit with this examiner.

Request for Continued Examination (RCE) Effectiveness

When applicants file an RCE with this examiner, 32.0% of applications are subsequently allowed. This success rate is in the 67% percentile among all examiners. Strategic Insight: RCEs show above-average effectiveness with this examiner. Consider whether your amendments or new arguments are strong enough to warrant an RCE versus filing a continuation.

After-Final Amendment Practice

This examiner enters after-final amendments leading to allowance in 17.3% of cases where such amendments are filed. This entry rate is in the 20% percentile among all examiners. Strategic Recommendation: This examiner rarely enters after-final amendments compared to other examiners. You should generally plan to file an RCE or appeal rather than relying on after-final amendment entry. Per MPEP § 714.12, primary examiners have discretion in entering after-final amendments, and this examiner exercises that discretion conservatively.

Pre-Appeal Conference Effectiveness

When applicants request a pre-appeal conference (PAC) with this examiner, 71.4% result in withdrawal of the rejection or reopening of prosecution. This success rate is in the 59% percentile among all examiners. Strategic Recommendation: Pre-appeal conferences show above-average effectiveness with this examiner. If you have strong arguments, a PAC request may result in favorable reconsideration.

Appeal Withdrawal and Reconsideration

This examiner withdraws rejections or reopens prosecution in 63.9% of appeals filed. This is in the 44% percentile among all examiners. Of these withdrawals, 30.4% occur early in the appeal process (after Notice of Appeal but before Appeal Brief). Strategic Insight: This examiner shows below-average willingness to reconsider rejections during appeals. Be prepared to fully prosecute appeals if filed.

Petition Practice

When applicants file petitions regarding this examiner's actions, 56.0% are granted (fully or in part). This grant rate is in the 58% percentile among all examiners. Strategic Note: Petitions show above-average success regarding this examiner's actions. Petitionable matters include restriction requirements (MPEP § 1002.02(c)(2)) and various procedural issues.

Examiner Cooperation and Flexibility

Examiner's Amendments: This examiner makes examiner's amendments in 0.0% of allowed cases (in the 10% percentile). This examiner rarely makes examiner's amendments compared to other examiners. You should expect to make all necessary claim amendments yourself through formal amendment practice.

Quayle Actions: This examiner issues Ex Parte Quayle actions in 2.2% of allowed cases (in the 69% percentile). This examiner issues Quayle actions more often than average when claims are allowable but formal matters remain (MPEP § 714.14).

Prosecution Strategy Recommendations

Based on the statistical analysis of this examiner's prosecution patterns, here are tailored strategic recommendations:

  • Plan for RCE after final rejection: This examiner rarely enters after-final amendments. Budget for an RCE in your prosecution strategy if you receive a final rejection.
  • Plan for extended prosecution: Applications take longer than average with this examiner. Factor this into your continuation strategy and client communications.

Relevant MPEP Sections for Prosecution Strategy

  • MPEP § 713.10: Examiner interviews - available before Notice of Allowance or transfer to PTAB
  • MPEP § 714.12: After-final amendments - may be entered "under justifiable circumstances"
  • MPEP § 1002.02(c): Petitionable matters to Technology Center Director
  • MPEP § 1004: Actions requiring primary examiner signature (allowances, final rejections, examiner's answers)
  • MPEP § 1207.01: Appeal conferences - mandatory for all appeals
  • MPEP § 1214.07: Reopening prosecution after appeal

Important Disclaimer

Not Legal Advice: The information provided in this report is for informational purposes only and does not constitute legal advice. You should consult with a qualified patent attorney or agent for advice specific to your situation.

No Guarantees: We do not provide any guarantees as to the accuracy, completeness, or timeliness of the statistics presented above. Patent prosecution statistics are derived from publicly available USPTO data and are subject to data quality limitations, processing errors, and changes in USPTO practices over time.

Limitation of Liability: Under no circumstances will IronCrow AI be liable for any outcome, decision, or action resulting from your reliance on the statistics, analysis, or recommendations presented in this report. Past prosecution patterns do not guarantee future results.

Use at Your Own Risk: While we strive to provide accurate and useful prosecution statistics, you should independently verify any information that is material to your prosecution strategy and use your professional judgment in all patent prosecution matters.