Commits · a6cee81c41f4390bbf87b3433bfc081aaab19a0e · Samuele Tosatto / Optimistic-DQN

May 09, 2018
- backup fixed · a6cee81c
  Samuele authored 7 years ago
  
  a6cee81c
May 08, 2018
- Backup improved. Final experiment setup for atari. · 8a712c78
  Samuele authored 7 years ago
  
  8a712c78
May 07, 2018
- exploiting ODQN · 46e1b1bf
  Samuele authored 7 years ago
  
  46e1b1bf
May 03, 2018
- Taxi experiment · 7aaa257d
  Samuele authored 7 years ago
  
  7aaa257d
May 01, 2018
- fix in net atari · aa0fb20d
  carloderamo authored 7 years ago
  
  aa0fb20d
Apr 30, 2018
- weighted policy option included · 827a54d2
  Samuele authored 7 years ago
  
  827a54d2
Apr 28, 2018
- bug fixing · f58ffe22
  Samuele authored 7 years ago
  
  f58ffe22
- Exception in case of backup missing · e875206b
  Samuele authored 7 years ago
  
  e875206b
Apr 27, 2018
- Pacman Experiment · 4f58b03c
  Samuele authored 7 years ago
  
  4f58b03c
- Backup + no autokill + evaluation default w eps= 0.005 · 66977f04
  Samuele authored 7 years ago
  
  66977f04
- montezuma no epsilon · 8f539412
  Samuele authored 7 years ago
  
  8f539412
Apr 26, 2018
- save eta · 7d93a4c3
  Samuele authored 7 years ago
  
  7d93a4c3
- Atari experiments · 9a3e7ad9
  Samuele authored 7 years ago
  
  9a3e7ad9
- adaptive ODQN for atari · 82467717
  Samuele authored 7 years ago
  
  82467717
- new breakout · fbfc4e8b
  Samuele authored 7 years ago
  
  fbfc4e8b
- learning rate · 80f4ff20
  Samuele authored 7 years ago
  
  80f4ff20
- modification of dqn' · e578e258
  Samuele authored 7 years ago
  
  e578e258
- export variables · 02047231
  Samuele authored 7 years ago
  
  02047231
- load cuda · 9e1f64ea
  Samuele authored 7 years ago
  
  9e1f64ea
- name and path of output changed · de8d9ed0
  Samuele authored 7 years ago
  
  de8d9ed0
- experiment w different names · d128a795
  Samuele authored 7 years ago
  
  d128a795
- MSE on simplenet; Adaptive DQN; experiment for CartPole to run on the server · 067c4ff1
  Samuele authored 7 years ago
  
  067c4ff1
Apr 25, 2018
- more robust entropy computation · a4f6ab08
  Samuele authored 7 years ago
  
  a4f6ab08
Apr 24, 2018
- AdaptiveOptimisticDQN · dcb9274d
  Samuele authored 7 years ago
  
  dcb9274d
- save and load replay memory · fba32fb8
  Samuele authored 7 years ago
  
  fba32fb8
- bug fixing · db42edbc
  Samuele authored 7 years ago
  
  db42edbc
- bug fix · 310dabe2
  Samuele authored 7 years ago
  
  310dabe2
- quiet · 3e45a850
  Samuele authored 7 years ago
  
  3e45a850
- bug fix · dd5f9c58
  Samuele authored 7 years ago
  
  dd5f9c58
- library setting · 8eb5dfdd
  Samuele authored 7 years ago
  
  8eb5dfdd
- mem spec updated · 80269eb2
  Samuele authored 7 years ago
  
  80269eb2
- optimistic vs bootstrapped · bd7d1282
  Samuele authored 7 years ago
  
  bd7d1282
- Atari setting · bbbdb125
  Samuele authored 7 years ago
  
  bbbdb125
- replay memory updated · 0553d98c
  carloderamo authored 7 years ago
  
  0553d98c
Apr 20, 2018
- bug corrected · 59623c8b
  Samuele authored 7 years ago
  
  59623c8b
- for cpu · e5e03d6b
  Samuele authored 7 years ago
  
  e5e03d6b
- Lunar Lander Experiment for the cluster · 37cb9c84
  Samuele authored 7 years ago
  
  37cb9c84
- Compute the bonus · 7fdc253f
  Samuele authored 7 years ago
  
  7fdc253f
- Experiments for gym environments · 266b3642
  Samuele authored 7 years ago
  
  266b3642
- Merge branch 'master' of git.ias.informatik.tu-darmstadt.de:tosatto/Optimistic-DQN · 3bf063b0
  Samuele authored 7 years ago
  3bf063b0