ACM JOURNAL ON EMERGING TECHNOLOGIES IN COMPUTING SYSTEMS | |
Fast Linear Interpolation | |
Article | |
Zhang, Nathan1,4  Canini, Kevin2,3  Silva, Sean3  Gupta, Maya2,3  | |
[1] Stanford Univ, Stanford, CA 94305 USA.;Google Res, Mountain View, CA USA.;Google Inc, 1600 Amphitheatre Pkwy, Mountain View, CA USA.;Gates Comp Sci Bldg,353 Jane Stanford Way, Stanford, CA 94305 USA. | |
关键词: Compiler; interpolation; | |
DOI : 10.1145/3423184 | |
来源: SCIE | |
【 摘 要 】
We present fast implementations of linear interpolation operators for piecewise linear functions and multidimensional 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 - 10x faster on a standard CPU compared to an optimized C++ interpreter implementation.
【 授权许可】
Free
【 预 览 】
Files | Size | Format | View |
---|---|---|---|
RO202303090297476ZK.pdf | 3281KB | download |
【 参考文献 】
- [1]
- [2]
- [3]
- [4]
- [5]
- [6]
- [7]
- [8]
- [9]
- [10]
- [11]
- [12]
- [13]
- [14]
- [15]
- [16]
- [17]
- [18]
- [19]
- [20]
- [21]
- [22]
- [23]
- [24]
- [25]
- [26]
- [27]
- [28]
- [29]
- [30]
- [31]
- [32]
- [33]
- [34]