Tensor Data Scattering and the Impossibility of Slicing Theorem
This work addresses the problem of efficiently representing and processing sparse tensors for deep learning practitioners and hardware designers, offering theoretical insights for performance optimization.
This paper proposes a standard representation for sparse tensors and establishes a theoretical framework for tensor data scattering methods. It introduces a theorem demonstrating the impossibility of slicing in tensor data scattering, which is crucial for performance analysis and accelerator optimization.
This paper proposes a standard way to represent sparse tensors. A broad theoretical framework for tensor data scattering methods used in various deep learning frameworks is established. This paper presents a theorem that is very important for performance analysis and accelerator optimization for implementing data scattering. The theorem shows how the impossibility of slicing happens in tenser data scattering. A sparsity measuring formula is provided, which can effectively indicate the storage efficiency of sparse tensor and the possibility of parallelly using it. The source code, including CUDA code, is provided in a related open-source project.