Matches in SemOpenAlex for { <https://semopenalex.org/work/W4288104003> ?p ?o ?g. }
Showing items 1 to 71 of
71
with 100 items per page.
- W4288104003 abstract "One of the key challenges arising when compilers vectorize loops for today's SIMD-compatible architectures is to decide if vectorization or interleaving is beneficial. Then, the compiler has to determine how many instructions to pack together and how many loop iterations to interleave. Compilers are designed today to use fixed-cost models that are based on heuristics to make vectorization decisions on loops. However, these models are unable to capture the data dependency, the computation graph, or the organization of instructions. Alternatively, software engineers often hand-write the vectorization factors of every loop. This, however, places a huge burden on them, since it requires prior experience and significantly increases the development time. In this work, we explore a novel approach for handling loop vectorization and propose an end-to-end solution using deep reinforcement learning (RL). We conjecture that deep RL can capture different instructions, dependencies, and data structures to enable learning a sophisticated model that can better predict the actual performance cost and determine the optimal vectorization factors. We develop an end-to-end framework, from code to vectorization, that integrates deep RL in the LLVM compiler. Our proposed framework takes benchmark codes as input and extracts the loop codes. These loop codes are then fed to a loop embedding generator that learns an embedding for these loops. Finally, the learned embeddings are used as input to a Deep RL agent, which determines the vectorization factors for all the loops. We further extend our framework to support multiple supervised learning methods. We evaluate our approaches against the currently used LLVM vectorizer and loop polyhedral optimization techniques. Our experiments show 1.29X-4.73X performance speedup compared to baseline and only 3% worse than the brute-force search on a wide range of benchmarks." @default.
- W4288104003 created "2022-07-28" @default.
- W4288104003 creator A5023309381 @default.
- W4288104003 creator A5035134864 @default.
- W4288104003 creator A5041920173 @default.
- W4288104003 creator A5055703000 @default.
- W4288104003 creator A5059294839 @default.
- W4288104003 creator A5087003952 @default.
- W4288104003 date "2019-09-20" @default.
- W4288104003 modified "2023-10-16" @default.
- W4288104003 title "NeuroVectorizer: End-to-End Vectorization with Deep Reinforcement Learning" @default.
- W4288104003 doi "https://doi.org/10.48550/arxiv.1909.13639" @default.
- W4288104003 hasPublicationYear "2019" @default.
- W4288104003 type Work @default.
- W4288104003 citedByCount "0" @default.
- W4288104003 crossrefType "posted-content" @default.
- W4288104003 hasAuthorship W4288104003A5023309381 @default.
- W4288104003 hasAuthorship W4288104003A5035134864 @default.
- W4288104003 hasAuthorship W4288104003A5041920173 @default.
- W4288104003 hasAuthorship W4288104003A5055703000 @default.
- W4288104003 hasAuthorship W4288104003A5059294839 @default.
- W4288104003 hasAuthorship W4288104003A5087003952 @default.
- W4288104003 hasBestOaLocation W42881040031 @default.
- W4288104003 hasConcept C108583219 @default.
- W4288104003 hasConcept C111919701 @default.
- W4288104003 hasConcept C127705205 @default.
- W4288104003 hasConcept C13280743 @default.
- W4288104003 hasConcept C150552126 @default.
- W4288104003 hasConcept C154945302 @default.
- W4288104003 hasConcept C169590947 @default.
- W4288104003 hasConcept C173608175 @default.
- W4288104003 hasConcept C185798385 @default.
- W4288104003 hasConcept C199360897 @default.
- W4288104003 hasConcept C205649164 @default.
- W4288104003 hasConcept C41008148 @default.
- W4288104003 hasConcept C41608201 @default.
- W4288104003 hasConcept C41681595 @default.
- W4288104003 hasConcept C80444323 @default.
- W4288104003 hasConcept C97541855 @default.
- W4288104003 hasConceptScore W4288104003C108583219 @default.
- W4288104003 hasConceptScore W4288104003C111919701 @default.
- W4288104003 hasConceptScore W4288104003C127705205 @default.
- W4288104003 hasConceptScore W4288104003C13280743 @default.
- W4288104003 hasConceptScore W4288104003C150552126 @default.
- W4288104003 hasConceptScore W4288104003C154945302 @default.
- W4288104003 hasConceptScore W4288104003C169590947 @default.
- W4288104003 hasConceptScore W4288104003C173608175 @default.
- W4288104003 hasConceptScore W4288104003C185798385 @default.
- W4288104003 hasConceptScore W4288104003C199360897 @default.
- W4288104003 hasConceptScore W4288104003C205649164 @default.
- W4288104003 hasConceptScore W4288104003C41008148 @default.
- W4288104003 hasConceptScore W4288104003C41608201 @default.
- W4288104003 hasConceptScore W4288104003C41681595 @default.
- W4288104003 hasConceptScore W4288104003C80444323 @default.
- W4288104003 hasConceptScore W4288104003C97541855 @default.
- W4288104003 hasLocation W42881040031 @default.
- W4288104003 hasOpenAccess W4288104003 @default.
- W4288104003 hasPrimaryLocation W42881040031 @default.
- W4288104003 hasRelatedWork W2021715972 @default.
- W4288104003 hasRelatedWork W2053732522 @default.
- W4288104003 hasRelatedWork W2090268225 @default.
- W4288104003 hasRelatedWork W2121941320 @default.
- W4288104003 hasRelatedWork W2378016289 @default.
- W4288104003 hasRelatedWork W2484588007 @default.
- W4288104003 hasRelatedWork W2487558162 @default.
- W4288104003 hasRelatedWork W2890419659 @default.
- W4288104003 hasRelatedWork W2983337431 @default.
- W4288104003 hasRelatedWork W3149597153 @default.
- W4288104003 isParatext "false" @default.
- W4288104003 isRetracted "false" @default.
- W4288104003 workType "article" @default.