Simon Rubanovich
99Patents
12h-index
63Co-inventors
83Inventor score
Filing activity: Dec 29, 2007 → Apr 5, 2024
Most-cited inventions
| Patent | Title | Area | Cited by | Status |
|---|---|---|---|---|
| US8984043B2 | Multiplying and adding matrices | Physics | 102 | Active |
| US11086623B2 | Systems, methods, and apparatuses for tile matrix multiplication and accumulation | Physics | 31 | Active |
| US9465580B2 | Math circuit for estimating a transcendental function | Physics | 21 | Active |
| US10866786B2 | Systems and methods for performing instructions to transpose rectangular tiles | Physics | 20 | Active |
| US11023235B2 | Systems and methods to zero a tile register pair | Physics | 19 | Active |
| US10963246B2 | Systems and methods for performing 16-bit floating-point matrix dot product instructions | Physics | 19 | Active |
| US11163565B2 | Systems, methods, and apparatuses for dot production operations | Physics | 19 | Active |
| US11093247B2 | Systems and methods to load a tile register pair | Physics | 19 | Active |
| US11175891B2 | Systems and methods to perform floating-point addition with selected rounding | Emerging Cross-Sectional Technologies | 17 | Active |
| US11200055B2 | Systems, methods, and apparatuses for matrix add, subtract, and multiply | Physics | 13 | Active |
| US11016731B2 | Using Fuzzy-Jbit location of floating-point multiply-accumulate results | Physics | 13 | Active |
| US10990397B2 | Apparatuses, methods, and systems for transpose instructions of a matrix operations accelerator | Physics | 13 | Active |
| US11669326B2 | Systems, methods, and apparatuses for dot product operations | Physics | 11 | Active |
| US9542154B2 | Fused multiply add operations using bit masks | Physics | 10 | Active |
| US11809869B2 | Systems and methods to store a tile register pair to memory | Physics | 9 | Active |
| US11816483B2 | Systems, methods, and apparatuses for matrix operations | Physics | 8 | Active |
| US9606770B2 | Multiply add functional unit capable of executing SCALE, ROUND, GETEXP, ROUND, GETMANT, REDUCE, RANGE and CLASS instructions | Physics | 7 | Active |
| US11972230B2 | Matrix transpose and multiply | Physics | 6 | Active |
| US9619226B2 | Systems, apparatuses, and methods for performing a horizontal add or subtract in response to a single instruction | Physics | 6 | Active |
| US11169802B2 | Systems, apparatuses, and methods for fused multiply add | Physics | 6 | Active |
| US11893389B2 | Systems and methods for performing 16-bit floating-point matrix dot product instructions | Physics | 5 | Active |
| US11941395B2 | Apparatuses, methods, and systems for instructions for 16-bit floating-point matrix dot product instructions | Physics | 5 | Active |
| US11403071B2 | Systems and methods for performing instructions to transpose rectangular tiles | Physics | 5 | Active |
| US11614936B2 | Systems and methods for performing 16-bit floating-point matrix dot product instructions | Physics | 5 | Active |
| US11609762B2 | Systems and methods to load a tile register pair | Physics | 5 | Active |
Source: USPTO / EPO open patent data. Inventor disambiguation is heuristic; counts are objective bibliographic measures.