提供可自由切换算法的DNN框架，轻松探索最优计算方案。
A Framework to Enable Algorithmic Design Choice Exploration in DNNs
- 支持细粒度算法控制，用户可自定义并替换DNN运算算法。
- 内置高性能实现，性能与PyTorch相当且无额外开销。
- 适合研究者快速验证新算法，提升DNN推理效率。
深度学习技术，尤其是深度神经网络（DNN），在多个领域取得了显著成功。这一成功伴随着支撑DNN运算的算法层面的重大进展与创新，这些改进算法具有极大提升DNN性能的潜力。然而，为DNN发现最佳算法并调整其使用这些算法的过程既困难又耗时。为此，我们提出一个开源框架，提供易于使用的细粒度算法控制能力，支持算法探索与选择。该框架内置了常见深度学习操作的高性能实现，使用户能够实现并选择自己的算法供DNN使用。实验表明，框架内置的加速实现输出结果与PyTorch等主流框架一致，性能相当，且不引入额外开销，因此性能完全取决于用户所选算法。
原文摘要 · Abstract (English)
Deep learning technologies, particularly deep neural networks (DNNs), have demonstrated significant success across many domains. This success has been accompanied by substantial advancements and innovations in the algorithms behind the operations required by DNNs. These enhanced algorithms hold the potential to greatly increase the performance of DNNs. However, discovering the best performing algorithm for a DNN and altering the DNN to use such algorithm is a difficult and time consuming task. To address this, we introduce an open source framework which provides easy to use fine grain algorithmic control for DNNs, enabling algorithmic exploration and selection. Along with built-in high performance implementations of common deep learning operations, the framework enables users to implement and select their own algorithms to be utilized by the DNN. The framework's built-in accelerated implementations are shown to yield outputs equivalent to and exhibit similar performance as implementations in PyTorch, a popular DNN framework. Moreover, the framework incurs no additional performance overhead, meaning that performance depends solely on the algorithms chosen by the user.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。