Login / Signup

PAME: precision-aware multi-exit DNN serving for reducing latencies of batched inferences.

Shulai ZhangWeihao CuiQuan ChenZhengnian ZhangYue GuanJingwen LengChao LiMinyi Guo
Published in: ICS (2022)
Keyphrases
  • high precision
  • website
  • three dimensional
  • high recall
  • recall rate
  • web services
  • artificial neural networks
  • training process