Issued Patents 2022
Showing 1–13 of 13 patents
| Patent # | Title | Co-Inventors | Date |
|---|---|---|---|
| 11507815 | Scaling half-precision floating point tensors for training deep neural networks | Naveen Mellempudi | 2022-11-22 |
| 11501139 | Scaling half-precision floating point tensors for training deep neural networks | Naveen Mellempudi | 2022-11-15 |
| 11494163 | Conversion hardware mechanism | Naveen Mellempudi, Chunhui Mei, Kristopher Wong, Dhiraj D. Kalamkar, Hong Jiang +2 more | 2022-11-08 |
| 11488008 | Hardware implemented point to point communication primitives for machine learning | Srinivas Sridharan, Karthikeyan Vaidyanathan | 2022-11-01 |
| 11468303 | Scaling half-precision floating point tensors for training deep neural networks | Naveen Mellempudi | 2022-10-11 |
| 11409525 | Apparatus and method for vector multiply and accumulate of packed words | Alexander Heinecke, Robert Valentine, Mark J. Charney | 2022-08-09 |
| 11373266 | Data parallelism and halo exchange for distributed machine learning | Karthikeyan Vaidyanathan, Srinivas Sridharan | 2022-06-28 |
| 11334796 | Optimized compute hardware for machine learning operations | Roger Gramunt, Mikhail Smelyanskiy, Jesus Corbal, Dheevatsa Mudigere, Naveen Mellempudi +1 more | 2022-05-17 |
| 11321805 | Dynamic precision management for integer deep learning primitives | Naveen Mellempudi, Dheevatsa Mudigere, Srinivas Sridharan | 2022-05-03 |
| 11321086 | Instructions for fused multiply-add operations with variable precision input operands | Naveen Mellempudi, Mrinmay Dutta, Arun Kumar, Dheevatsa Mudigere, Abhisek KUNDU | 2022-05-03 |
| 11314515 | Instructions and logic for vector multiply add with zero skipping | Supratim Pal, Sasikanth Avancha, Ishwar Bhati, Wei-Yu Chen, Ashutosh Garg +7 more | 2022-04-26 |
| 11275998 | Circuitry for low-precision deep learning | Martin Langhammer, Sudarshan Srinivasan, Gregg William Baeckler, Duncan Moss, Sasikanth Avancha | 2022-03-15 |
| 11270201 | Communication optimizations for distributed machine learning | Srinivas Sridharan, Karthikeyan Vaidyanathan, Chandrasekaran Sakthivel, Mikhail E. Smorkalov | 2022-03-08 |