PIT: Optimization of Dynamic Sparse Deep Learning Models via Permutation Invariant Transformation
Ningxin Zheng, Huiqiang Jiang, Quanlu Zhang, Zhenhua Han, Lingxiao Ma, Yuqing Yang, Fan Yang, Chengruidong Zhang, Lili Qiu, Mao Yang, Lidong Zhou
Abstract
Dynamic sparsity, where the sparsity patterns are unknown until runtime, poses a significant challenge to deep learning. The state-of-the-art sparsity-aware deep learning solutions are restricted to pre-defined, static sparsity patterns due to significant overheads associated with preprocessing. Efficient execution of dynamic sparse computation often faces the misalignment between the GPU-friendly tile configuration for efficient execution and the sparsity-aware tile shape that minimizes coverage wastes (non-zero values in tensor).
Topics & Concepts
Computer sciencePreprocessorDeep learningComputationPermutation (music)TileTransformation (genetics)Invariant (physics)Artificial intelligenceSparse matrixAlgorithmParallel computingMathematicsMathematical physicsArtGaussianVisual artsAcousticsQuantum mechanicsPhysicsChemistryBiochemistryGeneAdvanced Neural Network ApplicationsMultimodal Machine Learning ApplicationsAdvanced Image and Video Retrieval Techniques