| 12511516 |
Training neural networks based on dual pipeline architectures |
Mattheus Cornelis Antonius Adrianus Heddes, Torsten Hoefler, Kevin Colwell |
2025-12-30 |
|
| 12443848 |
Neural network activation compression with narrow block floating-point |
Daniel Lo, Eric S. Chung, Yinlin Zhao, Ritchie Zhao |
2025-10-14 |
|
| 12436801 |
Deep learning scheduler toolkit |
Saurabh Agarwal |
2025-10-07 |
|
| 12277502 |
Neural network activation compression with non-uniform mantissas |
Daniel Lo, Eric S. Chung, Yiren Zhao |
2025-04-15 |
|
| 12165038 |
Adjusting activation compression for neural network training |
Daniel Lo, Bita Darvish Rouhani, Eric S. Chung, Yiren Zhao, Ritchie Zhao |
2024-12-10 |
$225,255,000 |
| 12067495 |
Neural network activation compression with non-uniform mantissas |
Daniel Lo, Eric S. Chung, Yiren Zhao |
2024-08-20 |
$244,576,000 |
| 12056604 |
Highly performant pipeline parallel deep neural network training |
Vivek Seshadri, Deepak Narayanan, Aaron Harlap, Nikhil Devanur Rangarajan |
2024-08-06 |
$310,621,000 |
| 12045724 |
Neural network activation compression with outlier block floating-point |
Daniel Lo, Eric S. Chung, Yiren Zhao, Ritchie Zhao |
2024-07-23 |
$294,758,000 |
| 11868880 |
Mitigating communication bottlenecks during parameter exchange in data-parallel DNN training |
Nikhil Devanur Rangarajan, Jorgen Thelin, Guanhua Wang, Shivaram Venkataraman |
2024-01-09 |
$460,654,000 |
| 11715002 |
Efficient data encoding for deep neural network training |
Gennady Pekhimenko, Animesh Jain |
2023-08-01 |
$339,452,000 |
| 11562247 |
Neural network activation compression with non-uniform mantissas |
Daniel Lo, Eric S. Chung, Yiren Zhao |
2023-01-24 |
$221,394,000 |
| 10740195 |
Block storage by decoupling ordering from durability |
James W. Mickens, Vijaychidambaram Velayudhan Pillai |
2020-08-11 |
$116,675,000 |
| 10356187 |
IoT gateway for weakly connected settings |
Ranveer Chandra, Ashish Kapoor, Sudipta Narayan Sinha, Deepak Vasisht, Xinxin Jin +1 more |
2019-07-16 |
$67,164,000 |
| 10187292 |
Data center topology having multiple classes of reliability |
Monia Ghobadi, Ratul Mahajan, Danyang Zhuo, Xuan Zou |
2019-01-22 |
$62,390,000 |
| 10114709 |
Block storage by decoupling ordering from durability |
James W. Mickens, Vijaychidambaram Velayudhan Pillai |
2018-10-30 |
$61,995,000 |
| 10084868 |
IoT gateway for weakly connected settings |
Ranveer Chandra, Ashish Kapoor, Sudipta Narayan Sinha, Deepak Vasisht, Xinxin Jin +1 more |
2018-09-25 |
$110,824,000 |
| 10061791 |
Data management for connected devices |
Ratul Mahajan, Rayman Preet Singh, Trinabh Gupta, Jaeyeon Jung |
2018-08-28 |
$58,949,000 |
| 9798631 |
Block storage by decoupling ordering from durability |
James W. Mickens, Vijaychidambaram Velayudhan Pillai |
2017-10-24 |
$48,823,000 |
| 9342411 |
Reducing memory overhead of highly available, distributed, in-memory key-value caches |
Wendy A. Belluomini, Anna Povzner, Himabindu Pucha |
2016-05-17 |
$3,602,000 |
