| 12430546 |
General padding support for convolution on systolic arrays |
David Alexander Majnemer, Bjarke Hammersholt Roune |
2025-09-30 |
| 12399879 |
Approximate k nearest neighbors on hardware accelerators |
Felix Chern, Andrew Thomas Davis, Ruiqi Guo, Sanjiv Kumar, David Alexander Majnemer |
2025-08-26 |
| 12346782 |
Reshape and broadcast optimizations to avoid unnecessary data movement |
— |
2025-07-01 |
| 11907825 |
Training neural networks using distributed batch normalization |
Sameer Kumar |
2024-02-20 |
| 11763142 |
General padding support for convolution on systolic arrays |
David Alexander Majnemer, Bjarke Hammersholt Roune |
2023-09-19 |
| 11537939 |
Reshape and broadcast optimizations to avoid unnecessary data movement |
— |
2022-12-27 |
| 11500959 |
Multiple output fusion for operations performed in a multi-dimensional array of processing units |
David Alexander Majnemer |
2022-11-15 |
| 11449739 |
General padding support for convolution on systolic arrays |
David Alexander Majnemer, Bjarke Hammersholt Roune |
2022-09-20 |
| 9477599 |
Write combining cache microarchitecture for synchronization events |
Bradford M. Beckmann |
2016-10-25 |
| 9436395 |
Mechanisms to save user/kernel copy for cross device communications |
Shuai Che |
2016-09-06 |
| 9411652 |
Runtime for automatically load-balancing and synchronizing heterogeneous computer systems with scoped synchronization |
Derek Robert Hower |
2016-08-09 |
| 9396112 |
Hierarchical write-combining cache coherence |
Bradford M. Beckmann |
2016-07-19 |
| 9361118 |
Method for memory consistency among heterogeneous computer components |
Derek Robert Hower, Mark D. Hill, David A. Wood, Steven K. Reinhardt, Benedict R. Gaster +1 more |
2016-06-07 |
