Neural Architecture Search as Program Transformation Exploration

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Improving the performance of deep neural networks (DNNs) is important to both the compiler and neural architecture search (NAS) communities. Compilers apply program transformations in order to exploit hardware parallelism and memory hierarchy. However, legality concerns mean they fail to exploit the natural robustness of neural networks. In contrast, NAS techniques mutate networks by operations such as the grouping or bottlenecking of convolutions, exploiting the resilience of DNNs. In this work, we express such neural architecture operations as program transformations whose legality depends on a notion of representational capacity. This allows them to be combined with existing transformations into a unified optimization framework. This unification allows us to express existing NAS operations as combinations of simpler transformations. Crucially, it allows us to generate and explore new tensor convolutions. We prototyped the combined framework in TVM and were able to find optimizations across different DNNs, that significantly reduce inference time - over 3× in the majority of cases. Furthermore, our scheme dramatically reduces NAS search time.
Original languageEnglish
Title of host publicationProceedings of the 26th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS 2021)
PublisherACM Association for Computing Machinery
Pages915–927
Number of pages13
ISBN (Electronic)9781450383172
DOIs
Publication statusPublished - 19 Apr 2021
Event26th ACM International Conference on Architectural Support for Programming Languages and Operating Systems - https://asplos-conference.org/, Virtual, United States
Duration: 19 Apr 202123 Apr 2021

Conference

Conference26th ACM International Conference on Architectural Support for Programming Languages and Operating Systems
Abbreviated titleASPLOS 2021
CountryUnited States
CityVirtual
Period19/04/2123/04/21

Keywords

  • program transformations
  • neural networks

Fingerprint Dive into the research topics of 'Neural Architecture Search as Program Transformation Exploration'. Together they form a unique fingerprint.

Cite this