POSET-RL: Phase ordering for Optimizing Size and Execution Time using Reinforcement Learning

Jain, Shalini; Andaluri, Yashas; VenkataKeerthy, S.; Upadrasta, Ramakrishna

doi:10.1109/ispass55109.2022.00012

Cited by 4 publications

(1 citation statement)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent work [26] has identified two major problems in the application of machine learning to compiler optimization, 1) phase ordering problem and 2) selecting the best optimization. The most used machine learning approach recently is the reinforcement learning model like DQN(Deep Q learning) [27]in phase ordering problems. Autophase [28] and some LLVM phase ordering research [29] [30] had shown that well-decision phase ordering could enhance the performance and reduce the code size.…”

Section: Related Workmentioning

confidence: 99%

Accelerating Halide framework by automatic Affine scheduling using the stochastic optimization algorithm on MLIR

Wang,

Shen,

Liao

2023

Preprint

View full text Add to dashboard Cite

The increasing usage of computer vision algorithms in camera-centric devices has led to a growing need for optimizing these algorithms to improve their performance on resource-constrained platforms. Halide is a language specific to image processing algorithms that separates the algorithm's scheduling from its implementation, resulting in high performance. This thesis proposes an approach to improve the performance of Halide computer vision algorithms using stochastic algorithms such as simulated annealing to optimize scheduling, which enables exploring the global optimum of affine scheduling in constrained time. To convert the Halide program to MLIR, we use novel compile flows, namely the Halide to MLIR (HTM) converter. The efficacy of the approach will be evaluated on different platforms, such as x86, ARM, and RISC-V. The study demonstrates the potential of MLIR's transformation and optimization capabilities on Affine dialects and highlights the need for tuning infrastructure to fully leverage MLIR's optimization capabilities.

show abstract