LEGION
Official implementation of paper on Nature Machine Intelligence: "Preserving and Combining Knowledge in Robotic Lifelong Reinforcement Learning"
File Explorer
Download Latest Version (.zip)- config.yml
- CODE_OF_CONDUCT.md
- CONTRIBUTING.md
- ISSUE_TEMPLATE.md
- PULL_REQUEST_TEMPLATE.md
- continuouslearning_actor.yaml
- continuouslearning_critic.yaml
- continuouslearning_encoder.yaml
- continuouslearning_gradnorm.yaml
- continuouslearning_mask.yaml
- continuouslearning_multitask.yaml
- continuouslearning_task_to_encoder_cluster.yaml
- continuouslearning_transition_model.yaml
- gradnorm.yaml
- hipbmdp_actor.yaml
- hipbmdp_critic.yaml
- hipbmdp_encoder.yaml
- hipbmdp_gradnorm.yaml
- hipbmdp_mask.yaml
- hipbmdp_multitask.yaml
- hipbmdp_task_to_encoder_cluster.yaml
- hipbmdp_transition_model.yaml
- metaworld_actor.yaml
- metaworld_critic.yaml
- metaworld_encoder.yaml
- metaworld_gradnorm.yaml
- metaworld_mask.yaml
- metaworld_multitask.yaml
- metaworld_task_to_encoder_cluster.yaml
- metaworld_transition_model.yaml
- continuouslearning_actor.yaml
- continuouslearning_alpha.yaml
- continuouslearning_critic.yaml
- continuouslearning_decoder.yaml
- continuouslearning_encoder.yaml
- hipbmdp_actor.yaml
- hipbmdp_alpha.yaml
- hipbmdp_critic.yaml
- hipbmdp_decoder.yaml
- hipbmdp_encoder.yaml
- metaworld_actor.yaml
- metaworld_alpha.yaml
- metaworld_critic.yaml
- metaworld_decoder.yaml
- metaworld_encoder.yaml
- deepmdp.yaml
- distral.yaml
- gradnorm_deepmdp.yaml
- gradnorm_sac.yaml
- gradnorm_state_deepmdp.yaml
- gradnorm_state_sac.yaml
- hipbmdp.yaml
- pcgrad_deepmdp.yaml
- pcgrad_sac.yaml
- pcgrad_sac_ae.yaml
- pcgrad_state_deepmdp.yaml
- pcgrad_state_sac.yaml
- sac.yaml
- sac_ae.yaml
- sac_dpmm.yaml
- state_deepmdp.yaml
- state_sac.yaml
- dmcontrol-finger-spin-distribution-v0.yaml
- metaworld-ml1.yaml
- metaworld-mt1.yaml
- metaworld-mt10.yaml
- metaworld-mt50.yaml
- continuouslearning.yaml
- hipbmdp.yaml
- metaworld.yaml
- mtrl.yaml
- mtrl.yaml
- mtrl.yaml
- all.yaml
- continuouslearning.yaml
- hipbmdp.yaml
- metaworld.yaml
- mtrl.yaml
- continuouslearning.yaml
- hipbmdp.yaml
- metaworld.yaml
- __init__.py
- config.yaml
- latent_dpmm_subtask1.npz
- latent_dpmm_subtask1_tsne.csv
- latent_dpmm_subtask3.npz
- latent_dpmm_subtask3_tsne.csv
- latent_dpmm_subtask5.npz
- latent_dpmm_subtask5_tsne.csv
- latent_dpmm_subtask7.npz
- latent_dpmm_subtask7_tsne.csv
- latent_dpmm_subtask9.npz
- latent_dpmm_subtask9_tsne.csv
- latent_samples_step10000000_subtask9_1.npz
- latent_samples_step1000000_subtask0_1.npz
- latent_samples_step2000000_subtask1_1.npz
- latent_samples_step3000000_subtask2_1.npz
- latent_samples_step4000000_subtask3_1.npz
- latent_samples_step5000000_subtask4_1.npz
- latent_samples_step6000000_subtask5_1.npz
- latent_samples_step7000000_subtask6_1.npz
- latent_samples_step8000000_subtask7_1.npz
- latent_samples_step9000000_subtask8_1.npz
- legion_crl_eval_meanstd.csv
- legion_crl_eval_success1.csv
- legion_crl_eval_success2.csv
- legion_crl_eval_success3.csv
- legion_crl_eval_success4.csv
- legion_crl_eval_success5.csv
- legion_crl_eval_success6.csv
- legion_crl_eval_success7.csv
- legion_crl_smooth.csv
- 0_buffer_tsne.csv
- 10_buffer_tsne.csv
- 11_buffer_tsne.csv
- 12_buffer_tsne.csv
- 13_buffer_tsne.csv
- 14_buffer_tsne.csv
- 1_buffer_tsne.csv
- 2_buffer_tsne.csv
- 3_buffer_tsne.csv
- 4_buffer_tsne.csv
- 5_buffer_tsne.csv
- 6_buffer_tsne.csv
- 7_buffer_tsne.csv
- 8_buffer_tsne.csv
- 9_buffer_tsne.csv
- fewshot_task_0.csv
- fewshot_task_1.csv
- fewshot_task_4.csv
- fewshot_task_6.csv
- fewshot_task_9.csv
- latent_samples_step10000000_subtask9_1.npz
- latent_samples_step1000000_subtask0_1.npz
- latent_samples_step11000000_subtask0_1.npz
- latent_samples_step12000000_subtask1_1.npz
- latent_samples_step13000000_subtask4_1.npz
- latent_samples_step14000000_subtask6_1.npz
- latent_samples_step15000000_subtask9_1.npz
- latent_samples_step2000000_subtask1_1.npz
- latent_samples_step3000000_subtask4_1.npz
- latent_samples_step4000000_subtask6_1.npz
- latent_samples_step5000000_subtask9_1.npz
- latent_samples_step6000000_subtask0_1.npz
- latent_samples_step7000000_subtask1_1.npz
- latent_samples_step8000000_subtask4_1.npz
- latent_samples_step9000000_subtask6_1.npz
- button-press-topdown-v2.mp4
- door-open-v2.mp4
- drawer-close-v2.mp4
- faucet-open-v2.mp4
- peg-unplug-side-v2.mp4
- pick-place-v2-2.mp4
- push-v2.mp4
- reach-v2.mp4
- window-close-v2.mp4
- window-open-v2.mp4
- button-press-topdown-v2.mp4
- door-open-v2.mp4
- drawer-close-v2.mp4
- faucet-open-v2.mp4
- peg-unplug-side-v2.mp4
- pick-place-v2.mp4
- push-v2.mp4
- reach-v2.mp4
- window-close-v2.mp4
- window-open-v2.mp4
- rand_eval_1.log
- rand_eval_2.log
- rand_eval_3.log
- rand_eval_4.log
- rand_eval_5.log
- rev_eval_1.log
- rev_eval_2.log
- rev_eval_3.log
- rev_eval_4.log
- rev_eval_5.log
- all_models_multitask_eval_success.csv
- latent.csv
- latent_samples.npz
- legion_eval_success0.csv
- legion_eval_success1.csv
- legion_eval_success2.csv
- legion_eval_success3.csv
- legion_eval_success4.csv
- legion_eval_success5.csv
- actor_0.95.pt
- Screenshot from 2024-09-04 11-14-10.png
- task_encoder_0.95.pt
- vae_0.95.pt
- legion_eval.ipynb
- readme.md
- bulma-carousel.min.css
- bulma-slider.min.css
- bulma.css.map.txt
- bulma.min.css
- fontawesome.all.min.css
- index.css
- framework_deployment.png
- framework_train.png
- img1.png
- img2.png
- img3.png
- movie_cover.png
- tum_icon.ico
- bulma-carousel.js
- bulma-carousel.min.js
- bulma-slider.js
- bulma-slider.min.js
- fontawesome.all.min.js
- index.js
- demo1.mp4
- demo_coffee-v2.mp4
- .nojekyll
- index.html
- README.md
- metaworld-mt10-v2.json
- metaworld-mt10.json
- metaworld-mt50.json
- __init__.py
- actor.py
- base.py
- bnp_model.py
- critic.py
- decoder.py
- encoder.py
- hipbmdp_theta.py
- moe_layer.py
- reward_decoder.py
- soft_modularization.py
- task_encoder.py
- transition_model.py
- __init__.py
- mt_obs.py
- task_info.py
- __init__.py
- abstract.py
- deepmdp.py
- distral.py
- grad_manipulation.py
- gradnorm.py
- hipbmdp.py
- pcgrad.py
- sac.py
- sac_ae.py
- sac_dpmm.py
- utils.py
- wrapper.py
- __init__.py
- run.py
- __init__.py
- builder.py
- types.py
- vec_env.py
- __init__.py
- continuouslearning.py
- dmcontrol.py
- experiment.py
- metaworld.py
- multitask.py
- utils.py
- __init__.py
- checkpointable.py
- config.py
- types.py
- utils.py
- video.py
- __init__.py
- logger.py
- replay_buffer.py
- replay_buffer_reservoir.py
- base.txt
- dev.txt
- docs.txt
- nox.txt
- INSTALL.md
- TRAIN_EVAL.md
- __init__.py
- decoder_test.py
- hipbmdp_components_test.py
- metaworld_components_test.py
- reward_decoder_test.py
- softmodularization_components_test.py
- transition_model_test.py
- hipbmdp_test.py
- metaworld_test.py
- utils.py
- __init__.py
- test_envs.py
- __init__.py
- hipbmdp_test.py
- metaworld_test.py
- utils.py
- __init__.py
- .gitignore
- .pre-commit-config.yaml
- .pylintrc
- .yamllint
- LICENSE
- main.py
- noxfile.py
- README.md
- setup.cfg
- setup.py
- towncrier.toml
// repository documentation
Was this content helpful?
(0 ratings)