Commit Graph

Select branches

Hide Pull Requests

finding_nans

master

mlpack

patch-1

patch-2

patch-3

patch-4

patch-5

patch-6

patch-7

tensorboard_logging

Mono Color

• 347d0d22cb Fix a bug Shangtong Zhang 2017-11-05 18:58:14 -07:00
• 34c6ca7dd8 Update figures Shangtong Zhang 2017-11-05 00:07:38 -06:00
• 29c6ce7266 Update figures Shangtong Zhang 2017-11-05 00:07:07 -06:00
• 736825a383 Update README Shangtong Zhang 2017-11-05 00:02:15 -06:00
• e06d7810b7 Update continuous_action_network.py patch-4 Mike Clark 2017-11-05 13:02:52 +08:00
• 5ce9264156 entropy equation? Mike Clark 2017-11-05 12:57:27 +08:00
• 9ccf247dfb entropy equation? patch-3 Mike Clark 2017-11-05 12:57:23 +08:00
• cc1e1415c7 Merge pull request #5 from wassname/patch-2 Shangtong Zhang 2017-11-04 09:13:29 -06:00
• 79311f02e6 found NaN's for small std wassname 2017-11-04 17:05:13 +08:00
• deab7043fc add eps to avoid NaN patch-2 Mike Clark 2017-11-04 16:24:54 +08:00
• a565202d4a Merge branch 'master' into finding_nans wassname 2017-11-01 18:58:55 +08:00
• 662c0834f6 This helps me avoid NaN when I have small rewards wassname 2017-11-01 18:58:03 +08:00
• ac9aa60f30 temporary asserts to find NaNs wassname 2017-11-01 18:54:30 +08:00
• 9ef309696f Merge pull request #4 from wassname/patch-1 Shangtong Zhang 2017-10-31 21:01:10 -06:00
• 212ff0bc70 moving epsilon outside softplus patch-1 Mike Clark 2017-11-01 10:52:23 +08:00
• 55ae92619a Update README Shangtong Zhang 2017-10-28 11:08:25 -06:00
• 9615195f5e Upgrade to PyTorch v0.2.0 Shangtong Zhang 2017-10-28 11:06:20 -06:00
• a59f2d2911 Benchmark DDPG with Roboschool Shangtong Zhang 2017-10-27 23:09:24 -06:00
• af360bb9ea Support roboschool Shangtong Zhang 2017-10-27 14:23:54 -06:00
• e832e2e89a Update README Shangtong Zhang 2017-10-26 23:06:20 -06:00
• 6682937b54 Support python3 and pytorch 0.2 Shangtong Zhang 2017-10-26 23:01:57 -06:00
• 1b6de16b4f Fix a critical bug in DDPG Shangtong Zhang 2017-10-15 10:05:33 -06:00
• 3ed6f2a9db Log wall time Shangtong Zhang 2017-10-14 16:38:51 -06:00
• c04b2a395f Update README Shangtong Zhang 2017-10-10 23:02:28 -06:00
• 51f125c01e Improve normalizer for scalar Shangtong Zhang 2017-10-07 21:23:02 -06:00
• fae9a85f31 Major update Shangtong Zhang 2017-10-06 22:10:44 -06:00
• 8189a5d136 Update README Shangtong Zhang 2017-10-06 11:48:58 -06:00
• c5cbc10f94 Refactor DDPG Shangtong Zhang 2017-10-06 11:16:21 -06:00
• 8b9fd8d24f Code cleanup Shangtong Zhang 2017-10-05 21:42:27 -06:00
• dd2443c1c3 DPPO Shangtong Zhang 2017-10-05 21:40:42 -06:00
• e58938e3fe Support async PPO, which doesn't work Shangtong Zhang 2017-10-04 22:33:26 -06:00
• c773da7d08 Support shared stats for continuous A3C Shangtong Zhang 2017-10-04 20:58:42 -06:00
• 3935dfc428 Unifying networks for continuous A3C and PPO Shangtong Zhang 2017-10-04 11:03:02 -06:00
• 3adceb5284 Gaussian actor Shangtong Zhang 2017-10-04 10:15:28 -06:00
• 27f54ed420 Single thread PPO Shangtong Zhang 2017-10-04 09:33:35 -06:00
• e60e9feecb Refactor for HRA Shangtong Zhang 2017-08-29 22:10:13 -06:00
• d52182882a Implementation of hybrid reward architecture Shangtong Zhang 2017-08-29 21:52:26 -06:00
• 74742312eb Compare with mlpack a3c mlpack Shangtong Zhang 2017-08-24 23:03:09 -06:00
• 2b30c6e999 Update README Shangtong Zhang 2017-08-03 21:02:03 -06:00
• 44e8a782af Refactor DQN Shangtong Zhang 2017-08-02 14:55:04 -06:00
• b161200f0f Add BN layer Shangtong Zhang 2017-08-02 14:35:51 -06:00
• 5116733f22 DDPG Pendulum Shangtong Zhang 2017-08-01 10:52:14 -06:00
• 1be3b44999 Continuous A3C Shangtong Zhang 2017-07-31 22:53:09 -06:00
• 473304d9a7 Update README Shangtong Zhang 2017-07-30 13:36:15 -06:00
• bbfcbf76f4 Major reversion Shangtong Zhang 2017-07-30 13:24:04 -06:00
• ce504e2d0f Major refactor Shangtong Zhang 2017-07-26 18:18:49 -06:00
• fb71f51ea7 Refactor file structure Shangtong Zhang 2017-07-25 21:00:13 -06:00
• c79cdb18ea Refactor async agent Shangtong Zhang 2017-07-25 12:06:54 -06:00
• 8b700774d2 Update logger Shangtong Zhang 2017-07-22 19:03:07 -06:00
• 2e6f49db85 Fix failed softmax layer test Shangtong Zhang 2017-07-22 10:07:30 -06:00
• 709cde1d4b Introduce tensorboard Shangtong Zhang 2017-07-22 09:28:02 -06:00
• a145a01cb6 Minor refactor Shangtong Zhang 2017-07-20 20:47:51 -06:00
• 7b7d37694e Update README Shangtong Zhang 2017-07-19 23:19:11 -06:00
• 5447d64caa DDPG Shangtong Zhang 2017-07-19 23:11:50 -06:00
• 0b88546091 Fix bug Shangtong Zhang 2017-07-19 10:29:29 -06:00
• f88b565f6b Update figures for A3C Shangtong Zhang 2017-06-19 09:24:16 -06:00
• b181f109f5 Major refactor Shangtong Zhang 2017-06-18 20:51:27 -06:00
• 8ea108e4bb Finalize async methods Shangtong Zhang 2017-06-10 21:08:40 -06:00
• 9c73abfbff Major refactor to support LSTM layer Shangtong Zhang 2017-06-08 18:36:04 -06:00
• af17bbbf94 Tune parameters Shangtong Zhang 2017-06-05 15:23:41 -06:00
• a9d4a0fa72 New network for value based async methods Shangtong Zhang 2017-06-04 19:21:04 -06:00
• b8b7602b35 Update README Shangtong Zhang 2017-06-04 14:47:22 -06:00
• 1840438561 Update README Shangtong Zhang 2017-06-04 14:43:36 -06:00
• 20871f918f Update figures Shangtong Zhang 2017-06-04 14:19:37 -06:00
• 1aebec066c Update Readme Shangtong Zhang 2017-06-03 15:07:53 -06:00
• 353541cd32 Double DQN and Dueling DQN Shangtong Zhang 2017-06-03 15:03:17 -06:00
• 52918587be Update Readme Shangtong Zhang 2017-06-03 10:47:23 -06:00
• 83783c4b7c Add figures for DQN breakout Shangtong Zhang 2017-06-03 10:33:03 -06:00
• d13cfe6e00 Update Readme Shangtong Zhang 2017-06-03 09:49:40 -06:00
• c0936e148b Update Readme Shangtong Zhang 2017-06-03 09:45:53 -06:00
• d5b455b2dc Add figure for A3C Shangtong Zhang 2017-06-03 09:43:44 -06:00
• 0a14c2ed31 Minor style fix Shangtong Zhang 2017-06-02 22:34:30 -06:00
• 80b0e85aba Update Readme Shangtong Zhang 2017-06-02 22:33:47 -06:00
• 56547e192a Refactor async agents Shangtong Zhang 2017-06-02 22:15:23 -06:00
• 3659cbed3a Fix bug for async agents Shangtong Zhang 2017-05-31 15:43:17 -06:00
• 774768640d Update async agents Shangtong Zhang 2017-05-30 23:39:26 -06:00
• 8da13efc01 Optimize DQN Shangtong Zhang 2017-05-29 23:51:23 -06:00
• 4908202976 Fix a bug of DQN Shangtong Zhang 2017-05-29 14:01:19 -06:00
• db580795f3 Atari wrapper Shangtong Zhang 2017-05-28 21:46:33 -06:00
• 863be9865a Refactor atari task Shangtong Zhang 2017-05-26 20:30:53 -06:00
• d77cc46cbf Fix a bug of replay Shangtong Zhang 2017-05-26 15:18:45 -06:00
• 45a3f00d32 Optimize replay buffer Shangtong Zhang 2017-05-24 19:45:42 -06:00
• be9c5cdcdb Optimize replay buffer Shangtong Zhang 2017-05-21 14:37:45 -06:00
• ad8c8ad853 Pixel atari games with DQN Shangtong Zhang 2017-05-20 22:09:18 -06:00
• 7c4919d6ac A3C Shangtong Zhang 2017-05-13 13:48:18 -06:00
• b55e190512 N-Step Q-Learning and One-Step Sarsa Shangtong Zhang 2017-05-11 23:50:41 -06:00
• 73dfc5a7e4 Minor style change Shangtong Zhang 2017-05-11 21:31:00 -06:00
• 4786bb8990 Update Readme Shangtong Zhang 2017-05-11 21:30:04 -06:00
• 4ce59dc419 Refactor APIs Shangtong Zhang 2017-05-11 16:17:51 -06:00
• 6fb24c95ff threading Shangtong Zhang 2017-05-10 21:03:45 -06:00
• dde67bece1 MP Shangtong Zhang 2017-05-10 14:12:37 -06:00
• de6d52d86f Refactor interfaces Shangtong Zhang 2017-05-05 22:24:51 -06:00
• 13212ca2cb Update readme Shangtong Zhang 2017-05-04 14:06:40 -06:00
• 966e3ebd60 Replace tensorflow with pytorch Shangtong Zhang 2017-05-04 14:05:40 -06:00
• 3f13ebc711 Crossprop Shangtong Zhang 2017-05-03 19:48:03 -06:00
• bb0429a5da Add copyright declaration Shangtong Zhang 2017-04-20 15:20:23 -06:00
• e8b469b796 LunarLander task Shangtong Zhang 2017-04-20 15:07:53 -06:00
• 72ef60c076 modularized DQN Shangtong Zhang 2017-04-20 14:03:44 -06:00
• 9101ed2eb0 Initial commit Shangtong Zhang 2017-04-20 13:59:53 -06:00