We present fast implementations of linear interpolation operators for piecewise linear functions and multi-dimensional look-up tables. These operators are common for efficient transformations in image processing and are the core operations needed for lattice models like deep lattice networks, a popular machine learning function class for interpretable, shape-constrained machine learning. We present new strategies for an efficient compiler-based solution using MLIR to accelerate linear interpolation. For real-world machine-learned multi-layer lattice models that use multidimensional linear interpolation, we show these strategies run 5-10× faster on a standard CPU compared to an optimized C++ interpreter implementation.

中文翻译：

---

快速线性插值

我们提出了用于分段线性函数和多维查找表的线性插值算子的快速实现。这些算子对于图像处理中的有效转换很常见，并且是深度格子网络等格子模型所需的核心操作，这是一种流行的机器学习函数类，用于可解释的、形状受限的机器学习。我们提出了使用 MLIR 来加速线性插值的高效基于编译器的解决方案的新策略。对于使用多维线性插值的真实世界机器学习多层点阵模型，我们展示了这些策略在标准 CPU 上的运行速度比优化的 C++ 解释器实现快 5-10 倍。