Ars Technica

Sparse Matrix Multiplication

So I have some very large matrices with nearly all elements being zero. Are matrix multiplication libraries able to avoid having to multiply the zero elements? More importantly, would they at least ...
Semiconductor Engineering

SparseP: Towards Efficient Sparse Matrix Vector Multiplication on Real Processing-In-Memory Systems

“Several manufacturers have already started to commercialize near-bank Processing-In-Memory (PIM) architectures. Near-bank PIM architectures place simple cores close to DRAM banks and can yield ...
Nature

Sparse Matrix Computation and Optimization Techniques

Sparse matrices underpin a vast range of scientific and engineering applications, from finite element analysis and computational fluid dynamics to graph processing and machine learning. By encoding ...
Design-Reuse

A custom RISC-V vector instruction to accelerate structured-sparse matrix multiplications

A novel AI-acceleration paper presents a method to optimize sparse matrix multiplication for machine learning models, particularly focusing on structured sparsity. Structured sparsity involves a ...