| Procrustes: a dataflow and accelerator for sparse deep neural network training D Yang, A Ghasemazar, X Ren, M Golub, G Lemieux, M Lis 2020 53rd Annual IEEE/ACM International Symposium on Microarchitecture …, 2020 | 67 | 2020 |
| Full deep neural network training on a pruned weight budget M Lis, M Golub, G Lemieux Proceedings of Machine Learning and Systems 1, 252-263, 2019 | 27 | 2019 |
| Dropback: Continuous pruning during training M Golub, G Lemieux, M Lis arXiv preprint arXiv:1806.06949, 21, 2018 | 15 | 2018 |
| With shared microexponents, a little shifting goes a long way B Darvish Rouhani, R Zhao, V Elango, R Shafipour, M Hall, ... Proceedings of the 50th Annual International Symposium on Computer …, 2023 | 11 | 2023 |
| Microscaling data formats for deep learning BD Rouhani, R Zhao, A More, M Hall, A Khodamoradi, S Deng, ... arXiv preprint arXiv:2310.10537, 2023 | 5 | 2023 |
| Turbo training for deep neural networks R Zhao, BD Rouhani, ES Chung, DC Burger, M Golub US Patent App. 17/330,395, 2022 | 1 | 2022 |
| DropBack: continuous pruning during deep neural network training M Golub University of British Columbia, 2018 | 1 | 2018 |
| Model customization of transformers for improved efficiency M Mesmakhosroshahi, BD Rouhani, ES Chung, DC Burger, MT Golub US Patent App. 17/748,912, 2023 | | 2023 |
| Sparsity masking methods for neural network training MT Golub, BD Rouhani, ES Chung, DC Burger US Patent App. 17/657,112, 2023 | | 2023 |
| Microscaling Data Formats for Deep Learning B Darvish Rouhani, R Zhao, A More, M Hall, A Khodamoradi, S Deng, ... arXiv e-prints, arXiv: 2310.10537, 2023 | | 2023 |
| Data-aware model pruning for neural networks V Elango, BD Rouhani, ES Chung, DC Burger, M Golub US Patent App. 17/334,613, 2022 | | 2022 |
| System for training an artificial neural network M Golub, R Zhao, E Chung, D Burger, BD Rouhani, G Yang, N Fusi US Patent App. 17/163,299, 2022 | | 2022 |
| A system and method for scheduling computing tasks on a network of autonomous vehicles M Golub, D Ramamurthy, R Nunes, H Burgmeier WO Patent WO2022002648A1, 2022 | | 2022 |
| Maximilian and Mieszko Lis. 2019. Full deep neural network training on a pruned weight budget LG Golub SysML, 0 | | |
