4 ms·
> Leveraging this diversity, we adapted AlphaTensor to specifically find algorithms that are fast on a given hardware, such as Nvidia V100 GPU, and Google TPU v
by lawrenceyan 4y ago
> Leveraging this diversity, we adapted AlphaTensor to specifically find algorithms that are fast on a given hardware, such as Nvidia V100 GPU, and Google TPU v2. These algorithms multiply large matrices 10-20% faster than the commonly used algorithms on the same hardware, which showcases AlphaTensor’s flexibility in optimising arbitrary objectives.
10-20% performance improvement in matrix multiplications is pretty amazing[0]!
[0]: https://www.nature.com/articles/s41586-022-05172-4/figures/5 https://www.nature.com/articles/s41586-022-05172-4/figures/5
- gnufx 4y agoThat doesn't say exactly what was run and how. From measurements with nvblas, that difference could be dwarfed by the effect of tile size and pinning buffers or not.