Login / Signup

Efficient Tensor Slicing for Multicore NPUs using Memory Burst Modeling.

Rafael C. F. SousaByungmin JungJaehwa KwakMichael FrankGuido Araujo
Published in: SBAC-PAD (2021)
Keyphrases
  • high order
  • computing power
  • higher order
  • lightweight
  • limited memory
  • memory management
  • data streams
  • information processing
  • computationally expensive
  • efficient implementation
  • modeling method