News for hackers. Stuff that matters.
Re:Learning to Optimize Tensor Programs by namibj on Tuesday May 22, 2018 @03:35PM (#17128751)
Compare to polyhedral optimization [0], supported in GCC via Graphite [1] and in LLVM via Polly [2]. These have a lower ceiling than hand-optimized assembler, but they automate the tuning of how to nest the loops for maximum benefits due to the cache hierarchy. Considering their ability to do so for general purpose number crunching loops, they are rather nice, but there is still some integration work needed, especially for LLVM, as lack a nice place in the existing optimization pipeline.

[0]: https://en.wikipedia.org/wiki/Polytope_model [1]: https://gcc.gnu.org/wiki/Graphite [2]: https://gcc.gnu.org/wiki/Graphite

Re:Learning to Optimize Tensor Programs by sidereal on Tuesday May 22, 2018 @06:08PM (#17129810)
Polyhedral optimisation is cool (Facebook has been using it to great effect for ML kernels recently [1]), but it’s not the end of the story. It’s complementary to this paper, which seems to be about learning an effective and transferrable cost model to guide the optimisation process (you could use that learned cost model in a polyhedral optimiser).

[1]: https://arxiv.org/abs/1802.04730

Re:Learning to Optimize Tensor Programs by namibj on Tuesday May 22, 2018 @06:25PM (#17129930)
Uhh.... this sounds interesting. Maybe we'll see it soon-ish (<5y) in LLVM/GCC.
Re:Learning to Optimize Tensor Programs by jules on Wednesday May 23, 2018 @05:11AM (#17132500)
It already exists! Check out Polly for LLVM.