Mirrors_Framework/deepmind-research
1
0
Fork 0
mirror of https://github.com/google-deepmind/deepmind-research.git synced 2026-05-09 21:07:49 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity
191 Commits 79 Branches 0 Tags
c8fa65149905b5cc4539d6cffa9b82a229ac185b
Commit Graph

191 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Louise Deason 49def83d1d Adds the code for "gated_linear_networks" to the files release.bara.sky and README.md for public release on github. 
PiperOrigin-RevId: 338219746
 2020-10-21 09:16:25 +00:00
Arthur Guez c45af649a7 Fix Arxiv link. 
PiperOrigin-RevId: 338119053
 2020-10-20 22:59:38 +01:00
David Pfau b2b1386a4d Add citation info for GEOMANCER after NeurIPS acceptance 
PiperOrigin-RevId: 338045024
 2020-10-20 22:58:51 +01:00
Arthur Guez 245211f318 Initial release of himo. 
PiperOrigin-RevId: 337916498
 2020-10-19 23:53:53 +01:00
Victoria Krakovna fefa95eb1f Added flags and parameters for the UVFA approximation to relative reachability (for the future tasks paper) 
PiperOrigin-RevId: 336881122
 2020-10-13 16:57:49 +01:00
Victoria Krakovna 2e48a73ee4 add explanation and requirements for running the UVFA approximation for the future tasks paper 
PiperOrigin-RevId: 336880872
 2020-10-13 16:56:57 +01:00
Victoria Krakovna 0b9372d5e6 add UVFA approximation for relative reachability penalty for the future tasks paper 
PiperOrigin-RevId: 336851625
 2020-10-13 13:01:02 +01:00
Victoria Krakovna bc398d8004 Added functionality for running on new environment variants in the future task paper 
PiperOrigin-RevId: 336680745
 2020-10-13 13:00:11 +01:00
Peter Wirnsberger 1d86410c90 Release of "learned_free_energy_estimation". 
PiperOrigin-RevId: 336058713
 2020-10-12 16:05:37 +01:00
Charlie Nash c82e368e0e Makes run.sh executable 
PiperOrigin-RevId: 334373303
 2020-10-12 16:03:28 +01:00
Raphael Koster 58a3ac707d Fix repeated words in readme. 
PiperOrigin-RevId: 334196593
 2020-10-12 16:03:27 +01:00
Alexander Novikov 99976cfaa9 An example of training CRR agent with TPU support 
PiperOrigin-RevId: 334117027
 2020-10-12 16:03:27 +01:00
Jake VanderPlas 0e5237df2a Use jax.api.device_put_sharded() in place of private JAX APIs. 
PiperOrigin-RevId: 332514384
 2020-09-23 16:43:36 +01:00
Mehdi Mirza 1d763e0beb Update the link to the arxiv tech report of physics planning games. 
PiperOrigin-RevId: 332458733
 2020-09-23 16:43:12 +01:00
Mihaela Rosca fd75024d61 Fix scratchgan requirements and update README. 
PiperOrigin-RevId: 332257241
 2020-09-18 15:33:39 +01:00
Alvaro Sanchez-Gonzalez 30434080cc Minor README changes. 
PiperOrigin-RevId: 332027406
 2020-09-16 18:02:10 +01:00
Alvaro Sanchez-Gonzalez 603a238733 Fixes bug in integration test setup which was making travis fail. 
PiperOrigin-RevId: 332020695
 2020-09-16 18:01:41 +01:00
Alvaro Sanchez-Gonzalez 0d8c06196b Adding "Learning to Simulate Complex Physics with Graph Networks". 
PiperOrigin-RevId: 332003336
 2020-09-16 15:59:22 +01:00
Alvaro Sanchez-Gonzalez 15d063b1ae Minor README changes. 
PiperOrigin-RevId: 331572243
 2020-09-16 15:58:56 +01:00
Alvaro Sanchez-Gonzalez d1f410c717 Adding README for the open sourced code. 
PiperOrigin-RevId: 331568645
 2020-09-16 15:58:29 +01:00
Alvaro Sanchez-Gonzalez 94f98470b9 Making integration test pull an actual dataset and do a few steps of training and evaluation. 
PiperOrigin-RevId: 331524333
 2020-09-16 15:58:01 +01:00
Mehdi Mirza ef672a0db9 Adds physics_planning_games. 
PiperOrigin-RevId: 331167767
 2020-09-11 18:21:07 +01:00
Florent Altché 7e7255eed1 Export typing annotations when available. 
PiperOrigin-RevId: 328527159
 2020-09-11 18:18:51 +01:00
Thomas Keck 85187de3dc Fix python dependencies for glassy dynamics. 
PiperOrigin-RevId: 328493638
 2020-08-26 16:59:08 +01:00
Florent Altché 9bd09fcace Minor fix to type annotations. 
PiperOrigin-RevId: 328491975
 2020-08-26 16:58:44 +01:00
Florent Altché 486d8caf21 Update glass_dynamics to Python 3 imports. 
PiperOrigin-RevId: 328363319
 2020-08-26 16:55:54 +01:00
Florent Altché 8457046b2c Add checkpoints from the ablation study. 
PiperOrigin-RevId: 328023346
 2020-08-26 16:54:56 +01:00
Charlie Nash 22c3daff19 Adds PolyGen to public Deepmind Research Github repository 
PiperOrigin-RevId: 327802622
 2020-08-21 15:42:20 +01:00
Florent Altché 56031adaa4 release of Bootstrap Your Own Latent. 
PiperOrigin-RevId: 327631980
 2020-08-21 08:22:52 +00:00
Florent Altché 2314aa74d5 Update BYOL readme with checkpoints URL and note on batchnorm init. 
PiperOrigin-RevId: 327614354
 2020-08-21 08:22:27 +00:00
Zafarali Ahmed 58ee8555ed Fix typo in the word "reinforcement learning" 
PiperOrigin-RevId: 327006404
 2020-08-21 08:21:58 +00:00
Saran Tunyasuvunakool b89cc0f495 [catch_carry] Add DOI and article number to BibTeX citation. 
PiperOrigin-RevId: 327006370
 2020-08-21 08:21:32 +00:00
Florent Altché 63fa5e72d5 Fix initial convolution channels not multiplied by width_multiplier. 
PiperOrigin-RevId: 327005191
 2020-08-21 08:21:04 +00:00
Clara Huiyi Hu 923ad3cff0 This notebook illustrates the CIFAR-10 experiments in the paper: 
PiperOrigin-RevId: 326141025
 2020-08-21 08:16:49 +00:00
Daniel J. Mankowitz 84321ad894 Add correct link to RWRL paper. 
PiperOrigin-RevId: 326012834
 2020-08-11 15:00:26 +01:00
Saran Tunyasuvunakool 7a07dc8e47 Release of code and dataset accompanying the SIGGRAPH 2020 publication "Catch & Carry: Reusable Neural Controllers for Vision-Guided Whole-Body Tasks". 
PiperOrigin-RevId: 325790467
 2020-08-10 15:49:14 +01:00
Shaobo Hou d687ecc9fb Update data URL. 
PiperOrigin-RevId: 325116144
 2020-08-06 03:05:07 +01:00
Shaobo Hou 60550a5bc6 Add a colab for generating figures. 
Export training curves to file and fix some inconsistencies.

PiperOrigin-RevId: 324825810
 2020-08-06 03:04:41 +01:00
Sergio Gomez 99aaa6930a Move to dopamine-rl version 3.1.2 in RL Unplugged 
PiperOrigin-RevId: 324071731
 2020-08-03 09:16:16 +00:00
Louise Deason e67cf45868 Release of MEMO dataset 
PiperOrigin-RevId: 324000883
 2020-08-03 09:15:49 +00:00
Mehdi Mirza c4af01fa75 Add board games to physics_planning_games. 
PiperOrigin-RevId: 323799306
 2020-08-03 09:15:12 +00:00
Shaobo Hou a24bda5ed0 Add GPE/GPI experiments. 
PiperOrigin-RevId: 323750949
 2020-07-29 14:36:59 +01:00
Alexander Novikov 59c0cf5044 Fix colab links 
PiperOrigin-RevId: 323321348
 2020-07-29 14:36:20 +01:00
Saran Tunyasuvunakool 69d8db961e Replace "pip install dm_control[locomotion_mazes]" commands with "pip install dm_control". 
The [locomotion_mazes] extra has been deprecated in dm_control, and the labmaze dependency is now on the main dm_control package.

PiperOrigin-RevId: 322820779
 2020-07-23 19:02:45 +01:00
Sergio Gomez ed1b814008 Fix problem with gsutil command to copy Atari data to local folder 
PiperOrigin-RevId: 322759837
 2020-07-23 19:02:05 +01:00
Sergio Gomez 1ea4cc033c Add RL Unplugged data loading code and examples 
PiperOrigin-RevId: 321746296
 2020-07-22 16:49:30 +01:00
Sergio Gomez bd29e1b710 Add image and citation links to RL Unplugged README 
PiperOrigin-RevId: 321197757
 2020-07-15 12:50:10 +01:00
Louise Deason 8188882a82 release of RL Unplugged 
PiperOrigin-RevId: 320985969
 2020-07-13 17:48:24 +00:00
Charlie Nash fd4cb5a29b Adding code for PolyGen. 
PiperOrigin-RevId: 320927671
 2020-07-13 17:47:53 +00:00
David Pfau a499eb9be7 Changed "published" to "described" in README to reflect preprint status 
PiperOrigin-RevId: 318282884
 2020-07-13 17:47:02 +00:00