{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/hierarchical-skills-for-efficient-exploration","title":"Hierarchical Skills for Efficient Exploration","arxiv_id":"2110.10809","date":"2021-10-20","proceeding":"NeurIPS 2021 12","authors":["Jonas Gehring","Gabriel Synnaeve","Andreas Krause","Nicolas Usunier"],"abstract":"In reinforcement learning, pre-trained low-level skills have the potential to greatly facilitate exploration. However, prior knowledge of the downstream task is required to strike the right balance between generality (fine-grained control) and specificity (faster learning) in skill design. In previous work on continuous control, the sensitivity of methods to this trade-off has not been addressed explicitly, as locomotion provides a suitable prior for navigation tasks, which have been of foremost interest. In this work, we analyze this trade-off for low-level policy pre-training with a new benchmark suite of diverse, sparse-reward tasks for bipedal robots. We alleviate the need for prior knowledge by proposing a hierarchical skill learning framework that acquires skills of varying complexity in an unsupervised manner. For utilization on downstream tasks, we present a three-layered hierarchical learning algorithm to automatically trade off between general and specific skills as required by the respective task. In our experiments, we show that our approach performs this trade-off effectively and achieves better results than current state-of-the-art methods for end- to-end hierarchical reinforcement learning and unsupervised skill discovery. Code and videos are available at https://facebookresearch.github.io/hsd3 .","url_abs":"https://arxiv.org/abs/2110.10809v1","url_pdf":"https://arxiv.org/pdf/2110.10809v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"hierarchical-skills-for-efficient-exploration","repo_url":"https://github.com/facebookresearch/hsd3","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"continuous-control","task_name":"Continuous Control"},{"task_slug":"efficient-exploration","task_name":"Efficient Exploration"},{"task_slug":"hierarchical-reinforcement-learning","task_name":"Hierarchical Reinforcement Learning"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"specificity","task_name":"Specificity"},{"task_slug":"continuous-control","task_name":"continuous-control"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[{"slug":"bipedal-skills","name":"bipedal-skills","full_name":"Bipedal Skills Benchmark for Reinforcement Learning"}],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2110.10809","atlas_url":"https://app.syntology.ai/?focus=2110.10809","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2110.10809"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/facebookresearch/hsd3","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":4,"unverified":3},"by_repo_kind":{"official":{"samples":7,"ran":4,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":7,"samples":[{"code_sha256_prefix":"651e37cf065e2beb","entry":"dim_select","repo":"facebookresearch/hsd3","repo_kind":"official","path":"hucc/utils.py","file_url":"https://github.com/facebookresearch/hsd3/blob/HEAD/hucc/utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":false,"mcp_get_code":{"code_sha256":"651e37cf065e2beb"}},{"code_sha256_prefix":"2c62336d2cae657c","entry":"shorthand","repo":"facebookresearch/hsd3","repo_kind":"official","path":"hucc/models/blocks.py","file_url":"https://github.com/facebookresearch/hsd3/blob/HEAD/hucc/models/blocks.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":false,"mcp_get_code":{"code_sha256":"2c62336d2cae657c"}},{"code_sha256_prefix":"e8ded132e7c6844c","entry":"sorted_nicely","repo":"facebookresearch/hsd3","repo_kind":"official","path":"hucc/utils.py","file_url":"https://github.com/facebookresearch/hsd3/blob/HEAD/hucc/utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":false,"mcp_get_code":{"code_sha256":"e8ded132e7c6844c"}},{"code_sha256_prefix":"2a6d36b0268e0afa","entry":"th_flatten","repo":"facebookresearch/hsd3","repo_kind":"official","path":"hucc/spaces.py","file_url":"https://github.com/facebookresearch/hsd3/blob/HEAD/hucc/spaces.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":false,"mcp_get_code":{"code_sha256":"2a6d36b0268e0afa"}},{"code_sha256_prefix":"e80e1bd22d60cab7","entry":"make_optim","repo":"facebookresearch/hsd3","repo_kind":"official","path":"hucc/utils.py","file_url":"https://github.com/facebookresearch/hsd3/blob/HEAD/hucc/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":false,"mcp_get_code":{"code_sha256":"e80e1bd22d60cab7"}},{"code_sha256_prefix":"52c77fc411e98fda","entry":"th_unflatten","repo":"facebookresearch/hsd3","repo_kind":"official","path":"hucc/spaces.py","file_url":"https://github.com/facebookresearch/hsd3/blob/HEAD/hucc/spaces.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":false,"mcp_get_code":{"code_sha256":"52c77fc411e98fda"}},{"code_sha256_prefix":"6de08f4942acb9b8","entry":"video_encode","repo":"facebookresearch/hsd3","repo_kind":"official","path":"hucc/render.py","file_url":"https://github.com/facebookresearch/hsd3/blob/HEAD/hucc/render.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":false,"mcp_get_code":{"code_sha256":"6de08f4942acb9b8"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}