Leo Parker Dirac
51Patents
15h-index
103Co-inventors
83Inventor score
Filing activity: Jun 12, 2014 → Jun 23, 2023
Most-cited inventions
| Patent | Title | Area | Cited by | Status |
|---|---|---|---|---|
| US10311371B1 | Machine learning based content delivery | Physics | 128 | Active |
| US10057227B1 | Determination of authentication mechanism | Electricity | 100 | Active |
| US10225365B1 | Machine learning based content delivery | Electricity | 64 | Active |
| US10311372B1 | Machine learning based content delivery | Electricity | 44 | Active |
| US10102480B2 | Machine learning service | Physics | 42 | Active |
| US10452992B2 | Interactive interfaces for machine learning model evaluations | Physics | 37 | Active |
| US10713589B1 | Consistent sort-based record-level shuffling of machine learning data | Physics | 34 | Active |
| US10366053B1 | Consistent randomized record-level splitting of machine learning data | Physics | 27 | Active |
| US9672474B2 | Concurrent binning of machine learning data | Physics | 26 | Active |
| US9881226B1 | Object relation builder | Physics | 26 | Active |
| US9886670B2 | Feature processing recipes for machine learning | Physics | 23 | Active |
| US10257275B1 | Tuning software execution environments using Bayesian models | Physics | 20 | Active |
| US11429762B2 | Simulation orchestration for training reinforcement learning models | Physics | 18 | Active |
| US10410224B1 | Determining item feature information from user content | Physics | 17 | Active |
| US10621019B1 | Using a client to manage remote machine learning jobs | Physics | 15 | Active |
| US10169715B2 | Feature processing tradeoff management | Electricity | 14 | Active |
| US10635973B1 | Recommendation system using improved neural network | Electricity | 13 | Active |
| US10970629B1 | Encodings for reversible sparse dimensionality reduction | Physics | 12 | Active |
| US11170309B1 | System for routing machine learning model inferences | Electricity | 12 | Active |
| US10540608B1 | Dynamically scaled training fleets for machine learning | Physics | 10 | Active |
| US11422863B2 | Attached accelerator scaling | Physics | 9 | Active |
| US10339465B2 | Optimized decision tree based models | Physics | 9 | Active |
| US10963810B2 | Efficient duplicate detection for machine learning data sets | Physics | 7 | Active |
| US10540606B2 | Consistent filtering of machine learning data | Physics | 6 | Active |
| US11100420B2 | Input processing for machine learning | Physics | 6 | Active |
Source: USPTO / EPO open patent data. Inventor disambiguation is heuristic; counts are objective bibliographic measures.