Login / Signup

Reward Only Training of Encoder-Decoder Digit Recognition Systems Based on Policy Gradient Methods.

Yilong PengHayato ShibataTakahiro Shinozaki
Published in: APSIPA (2018)
Keyphrases
  • digit recognition
  • reinforcement learning
  • video codec
  • decision trees
  • video sequences
  • training set
  • multi modal
  • policy gradient
  • decoding process
  • policy gradient methods