I wonder if Taco (The Tensor Algebra Compiler) might be useful technology for handling HTM’s sparse data representations:
New System for Performing “Tensor Algebra” Offers Faster Big-Data Analysis
The Tensor Algebra Compiler
-r
I wonder if Taco (The Tensor Algebra Compiler) might be useful technology for handling HTM’s sparse data representations:
New System for Performing “Tensor Algebra” Offers Faster Big-Data Analysis
The Tensor Algebra Compiler
-r
Please share your ideas why the Tensor Algebra Compiler might be useful for HTM.
I have no real knowledge of either HTM or the Tensor Algebra Compiler, but my understanding is that HTM implementations do a lot of work with sparse representations of data. This is the sort of thing that Taco is intended to address, albeit for a rather different use case.
-r