{"id":{"repo_id":"mit","oai_identifier":"oai:dspace.mit.edu:1721.1/162720"},"canonical_url":"https://search.dev.ndltd.org/etd/mit/oai:dspace.mit.edu:1721.1/162720","repository":{"repo_id":"mit","name":"MIT","base_url":"https://dspace.mit.edu/oai/request"},"display":{"title":"The Limits of Temporal Abstractions for Reinforcement Learning with Sparse Rewards","abstract":"Skills are temporal abstractions that are intended to improve reinforcement learning (RL) performance through hierarchical RL. Despite our intuition about the properties of an environment that make skills useful, there has been little theoretical work aimed to characterize these properties precisely. This work studies the utility of skills in sparse-reward environments with a discrete state space and finite action space. We show, both theoretically and empirically, that RL performance gains from skills are worse in environments where successful trajectories are less compressible. In environments with a highly incompressible distribution of successful trajectories, using unexpressive skills such as macroactions will provably worsen RL performance. We hope our findings can guide research on automatic skill discovery and help RL practitioners better decide when and how to use skills.","abstract_html":"Skills are temporal abstractions that are intended to improve reinforcement learning (RL) performance through hierarchical RL. Despite our intuition about the properties of an environment that make skills useful, there has been little theoretical work aimed to characterize these properties precisely. This work studies the utility of skills in sparse-reward environments with a discrete state space and finite action space. We show, both theoretically and empirically, that RL performance gains from skills are worse in environments where successful trajectories are less compressible. In environments with a highly incompressible distribution of successful trajectories, using unexpressive skills such as macroactions will provably worsen RL performance. We hope our findings can guide research on automatic skill discovery and help RL practitioners better decide when and how to use skills.","abstract_has_math":false,"creators":["Li, Zhening"],"institution":"Massachusetts Institute of Technology","degree_name":"Master","degree_level":null,"degree_discipline":null,"degree_department":"Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science","school":null,"contributors":[],"advisors":["Solar-Lezama, Armando"],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025-05","date_published":"2025-05","updated_at":"2026-07-22T22:22:28Z","subjects":[],"languages":[],"rights":["In Copyright - Educational Use Permitted","Copyright retained by author(s)"],"rights_urls":["https://rightsstatements.org/page/InC-EDU/1.0/"],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/1721.1/162720","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Solar-Lezama, Armando"]},{"key":"dc:contributor.department","label":"Department","values":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science"]},{"key":"dc:creator","label":"Author","values":["Li, Zhening"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2025-09-18T14:28:59Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2025-09-18T14:28:59Z"]},{"key":"dc:date.issued","label":"Date","values":["2025-05"]},{"key":"dc:publisher","label":"Institution","values":["Massachusetts Institute of Technology"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master","Master of Engineering in Electrical Engineering and Computer Science"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["In Copyright - Educational Use Permitted","Copyright retained by author(s)"]},{"key":"dc:rights.uri","label":"Rights URI","values":["https://rightsstatements.org/page/InC-EDU/1.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1721.1/162720"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Skills are temporal abstractions that are intended to improve reinforcement learning (RL) performance through hierarchical RL. Despite our intuition about the properties of an environment that make skills useful, there has been little theoretical work aimed to characterize these properties precisely. This work studies the utility of skills in sparse-reward environments with a discrete state space and finite action space. We show, both theoretically and empirically, that RL performance gains from skills are worse in environments where successful trajectories are less compressible. In environments with a highly incompressible distribution of successful trajectories, using unexpressive skills such as macroactions will provably worsen RL performance. We hope our findings can guide research on automatic skill discovery and help RL practitioners better decide when and how to use skills."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["M.Eng."]},{"key":"dc:title","label":"Title","values":["The Limits of Temporal Abstractions for Reinforcement Learning with Sparse Rewards"]}]}],"canonical_facts":{"dc:contributor.advisor":["Solar-Lezama, Armando"],"dc:contributor.department":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science"],"dc:creator":["Li, Zhening"],"dc:date.accessioned":["2025-09-18T14:28:59Z"],"dc:date.available":["2025-09-18T14:28:59Z"],"dc:date.issued":["2025-05"],"dc:description.abstract":["Skills are temporal abstractions that are intended to improve reinforcement learning (RL) performance through hierarchical RL. Despite our intuition about the properties of an environment that make skills useful, there has been little theoretical work aimed to characterize these properties precisely. This work studies the utility of skills in sparse-reward environments with a discrete state space and finite action space. We show, both theoretically and empirically, that RL performance gains from skills are worse in environments where successful trajectories are less compressible. In environments with a highly incompressible distribution of successful trajectories, using unexpressive skills such as macroactions will provably worsen RL performance. We hope our findings can guide research on automatic skill discovery and help RL practitioners better decide when and how to use skills."],"dc:description.degree":["M.Eng."],"dc:identifier.uri":["https://hdl.handle.net/1721.1/162720"],"dc:publisher":["Massachusetts Institute of Technology"],"dc:rights":["In Copyright - Educational Use Permitted","Copyright retained by author(s)"],"dc:rights.uri":["https://rightsstatements.org/page/InC-EDU/1.0/"],"dc:title":["The Limits of Temporal Abstractions for Reinforcement Learning with Sparse Rewards"],"dc:type":["Thesis"],"thesis:degree_name":["Master","Master of Engineering in Electrical Engineering and Computer Science"]},"updated_at":"2026-07-22T22:22:28Z"}