Login / Signup

ViT: Mixture-of-Experts Vision Transformer for Efficient Multi-task Learning with Model-Accelerator Co-design.

Hanxue LiangZhiwen FanRishov SarkarZiyu JiangTianlong ChenKai ZouYu ChengCong HaoZhangyang Wang
Published in: CoRR (2022)
Keyphrases
  • theoretical analysis
  • multi task learning
  • high order
  • gaussian processes
  • similarity measure
  • maximum likelihood
  • bayesian framework
  • bayesian model
  • multitask learning