On The Efficiency of Sparse-Tiled Tensor Graph Processing For Low Memory Usage
Antonio Cipolletta, Andrea Calimera
Abstract
The memory space taken to host and process large tensor graphs is a limiting factor for embedded ConvNets. Even though many data-driven compression pipelines have proven their efficacy, this work shows there is still room for optimization at the intersection with compute-oriented optimizations. We demonstrate that tensor pruning via weight sparsification can cooperate with a model-agnostic tiling strategy, leading ConvNets towards a new feasible region of the solution space. The collected results show for the first time fast versions of MobileNets deployed at full scale on an ARM M7 core with 512KB of RAM and 2MB of FLASH memory.
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Builds on2
Related papers
- Register Tiling for Unstructured Sparsity in Neural Network InferenceLucas Wilkinson, Kazem Cheshmi, Maryam Mehri DehnaviPLDI 2023 · 17 citations
- Convolutional Neural Network Compression through Generalized Kronecker Product DecompositionMarawan Gamal Abdel Hameed, Marzieh S. Tahaei, Ali Mosleh, Vahid Partovi NiaAAAI 2022 · 33 citations
- High Performance Depthwise and Pointwise Convolutions on Mobile DevicesPengfei Zhang, Eric Lo, Baotong LuAAAI 2020 · 54 citations
- Accelerating sparse DNN models without hardware-support via tile-wise sparsityCong Guo, Bo Yang Hsueh, Jingwen Leng, Yuxian Qiu et al.SC 2020 · 65 citations
- TileQ: Efficient Low-Rank Quantization of Mixture-of-Experts with 2D TilingHongyaoxing Gu, Xinzhe Chen, LIJUAN HU, Liu fangfangICML 2026
