Issued Patents All Time
Showing 25 most recent of 87 patents
| Patent # | Title | Co-Inventors | Date |
|---|---|---|---|
| 12430110 | Global modulo allocation in neural network compilation | Hongbin Zheng, Robert Geva | 2025-09-30 |
| 12400137 | Bidirectional network on a data-flow centric processor | Vignesh Vivekraja, Jiading Gai, Thiam Khean Hah | 2025-08-26 |
| 12373691 | Executing sublayers of a fully-connected layer | Ron Diamant | 2025-07-29 |
| 12321849 | Performing hardware operator fusion | Animesh Jain, Tobias Edler Von Koch, Yizhi Liu, Taemin Kim, Jindrich Zejda +3 more | 2025-06-03 |
| 12314837 | Multi-memory on-chip computational network | Ron Diamant | 2025-05-27 |
| 12205013 | Accelerated convolution of neural networks | Thiam Khean Hah, Richard John Heaton, Ron Diamant, Vignesh Vivekraja | 2025-01-21 |
| 12198041 | Efficient utilization of processing element array | Jeffrey T. Huynh, Ron Diamant, Hongbin Zheng, Yizhi Liu, Animesh Jain +5 more | 2025-01-14 |
| 12182695 | Fine-grained sparsity computations in systolic array | Paul Gilbert Meyer, Thiam Khean Hah, Ron Diamant, Vignesh Vivekraja | 2024-12-31 |
| 12182688 | Hierarchical partitioning of operators | Animesh Jain, Yizhi Liu, Hongbin Zheng, Jeffrey T. Huynh, Haichen Li +5 more | 2024-12-31 |
| 12159218 | Dropout layer in a neural network processor | Jiading Gai, Hongbin Zheng, Animesh Jain, Vignesh Vivekraja | 2024-12-03 |
| 12130885 | Emulating fine-grained sparsity in a systolic array | Paul Gilbert Meyer, Thiam Khean Hah, Ron Diamant, Vignesh Vivekraja | 2024-10-29 |
| 12106222 | Neural network training under memory restraint | Sudipta Sengupta, Ron Diamant, Vignesh Vivekraja | 2024-10-01 |
| 12093806 | Static memory allocation for neural network inference | Jindrich Zejda, Ron Diamant, Jeffrey T. Huynh, Drazen Borkovic, Richard John Heaton | 2024-09-17 |
| 12093801 | Neural network processing based on subgraph recognition | Richard John Heaton, Ron Diamant | 2024-09-17 |
| 12079734 | Compilation time reduction for memory and compute bound neural networks | Hongbin Zheng, Richard John Heaton | 2024-09-03 |
| 12073199 | Reducing computation in neural networks using self-modifying code | Vignesh Vivekraja, Yu Zhou, Ron Diamant, Richard John Heaton | 2024-08-27 |
| 12067492 | Processing for multiple input data sets in a multi-layer neural network | Dana Michelle Vantrease, Ron Diamant, Thomas A. Volpe | 2024-08-20 |
| 12008469 | Acceleration of neural networks with stacks of convolutional layers | Thiam Khean Hah, Richard John Heaton, Ron Diamant, Vignesh Vivekraja | 2024-06-11 |
| 12008466 | Processor with control flow | Ron Diamant, Thomas A. Volpe | 2024-06-11 |
| 11960997 | Circuit architecture with biased randomization | Ron Diamant | 2024-04-16 |
| 11948352 | Speculative training using partial gradients update | Patricio Kaplan | 2024-04-02 |
| 11941528 | Neural network training in a distributed system | Vignesh Vivekraja, Thiam Khean Hah, Ron Diamant, Richard John Heaton | 2024-03-26 |
| 11868878 | Executing sublayers of a fully-connected layer | Ron Diamant | 2024-01-09 |
| 11868875 | Data selection circuit | Ron Diamant, Jeffrey T. Huynh, Sundeep Amirineni | 2024-01-09 |
| 11868895 | Dynamic processing element array expansion | Ron Diamant, Richard John Heaton | 2024-01-09 |