Abstract | ||
---|---|---|
To accelerate DNNs inference, low-rank approximation has been widely adopted because of its solid theoretical rationale and efficient implementations. Several previous works attempted to directly approximate a pre-trained model by low-rank decomposition; however, small approximation errors in parameters can ripple over a large prediction loss. Apparently, it is not optimal to separate low-rank app... |
Year | DOI | Venue |
---|---|---|
2018 | 10.1109/EMC2-NIPS53020.2019.00011 | 2019 Fifth Workshop on Energy Efficient Machine Learning and Cognitive Computing - NeurIPS Edition (EMC2-NIPS) |
Keywords | DocType | Volume |
low-rank,decomposition,acceleration,pruning | Journal | abs/1812.02402 |
Issue | ISBN | Citations |
1 | 978-1-6654-2418-9 | 2 |
PageRank | References | Authors |
0.36 | 25 | 9 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yuhui Xu | 1 | 12 | 5.00 |
Yuxi Li | 2 | 81 | 15.02 |
Shuai Zhang | 3 | 45 | 6.63 |
Wei Wen | 4 | 353 | 18.09 |
Botao Wang | 5 | 171 | 77.07 |
Y Qi | 6 | 130 | 19.75 |
Yiran Chen | 7 | 3344 | 259.09 |
Weiyao Lin | 8 | 732 | 68.05 |
Hongkai Xiong | 9 | 512 | 82.84 |