junhyukoh/self-imitation-learning

Issues

• Calculating returns with signed rewards

  #7 opened 4 years ago by backpropper
  2

• How the policy and the value function use the same parameters $\theta$ ?

  #8 opened 4 years ago by xiaobanni
  1

• SIL Value update

  #3 opened 5 years ago by boscotsang
  2

• entropy in SIL policy loss

  #6 opened 6 years ago by gabrieledcjr
  0

• Policy 'lstm' doesn't work

  #5 opened 6 years ago by HaozhengLi
  1

• Key-Door-Treasure

  #4 opened 6 years ago by anagorko
  0

• np.sign(rewards)

  #2 opened 6 years ago by bhairavmehta95
  2