Combining Precision Tuning and Rewriting

Abstract

Precision tuning and rewriting can improve both the accuracy and speed of floating point expressions, yet these techniques are typically applied separately. This paper explores how finer-grained interleaving of precision tuning and rewriting can help automatically generate a richer set of Pareto-optimal accuracy versus speed trade-offs.

We introduce Pherbie (Pareto Herbie), a tool providing both precision tuning and rewriting, and evaluate interleaving these two strategies at different granularities. Our results demonstrate that finer-grained interleavings improve both the Pareto curve of candidate implementations and overall optimization time. On a popular set of tests from the FPBench suite, Pherbie finds both implementations that are significantly more accurate for a given cost and significantly faster for a given accuracy bound compared to baselines using precision tuning and rewriting alone or in sequence.

Talk

ARITH 2021 talk by Brett Saiki and Oliver Flatt.

BibTeX

@inproceedings{2021-arith-herbie,
  title     = {Combining Precision Tuning and Rewriting},
  author    = {Saiki, Brett and Flatt, Oliver and Nandi, Chandrakana and Panchekha, Pavel and Tatlock, Zachary},
  series    = {ARITH 2021},
  booktitle = {28th International Symposium on Computer Arithmetic},
  date      = {2021},
  doi       = {10.1109/ARITH51176.2021.00013},
  publisher = {IEEE Computer Society},
}

📝 publications index