Presentation
SSpMV: A Sparsity-aware SpMV Framework Empowered by Multimodal Machine Learning
DescriptionSparse Matrix-Vector Multiplication (SpMV) is an essential sparse operation in scientific computing and artificial intelligence.
Efficiently adapting SpMV algorithms to diverse matrices and architectures requires a framework capable of accurately recognizing sparse patterns and selecting the optimal implementation.
In this work, we introduce Sparsity-aware SpMV (SSpMV), a framework that integrates expert-designed features with multimodal representations to adaptively predict the best-performing algorithm and parameters.
For this purpose, we design a multimodal neural network called MM-Adapter, to capture diverse modalities to represent the computational features of SpMV.
Experimental results demonstrate that MM-Adapter achieves the highest accuracy of 81.05%, outperforming existing SpMV prediction models.
Furthermore, SSpMV consistently delivers substantial performance improvements over state-of-the-art sparse libraries across various multi-core platforms.
Efficiently adapting SpMV algorithms to diverse matrices and architectures requires a framework capable of accurately recognizing sparse patterns and selecting the optimal implementation.
In this work, we introduce Sparsity-aware SpMV (SSpMV), a framework that integrates expert-designed features with multimodal representations to adaptively predict the best-performing algorithm and parameters.
For this purpose, we design a multimodal neural network called MM-Adapter, to capture diverse modalities to represent the computational features of SpMV.
Experimental results demonstrate that MM-Adapter achieves the highest accuracy of 81.05%, outperforming existing SpMV prediction models.
Furthermore, SSpMV consistently delivers substantial performance improvements over state-of-the-art sparse libraries across various multi-core platforms.
Event Type
Research Manuscript
TimeTuesday, June 243:30pm - 3:45pm PDT
Location3008, Level 3
SYS3: Embedded Software