AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Hierarchical Reinforcement Learning with Timed Subgoals

NeurIPS 2021 · Reinforcement learning › Visual and representation RL · approach: Subgoal generation for hierarchical RL · 11 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use timed subgoals with time-interval conditioning (HiTS) instead of HACtwo-level hierarchy · Platformssuccess rate: 40% -> 86%Helped
augment HAC observation and subgoal space with absolute time (ablation)HAC · Platformssuccess rateHurt
HiTS vs HAC and SAC on Tennis2Dtwo-level hierarchy · Tennis2Dsuccess rate: around 40% within 20 million stepsHelped
HiTS vs HAC on static benchmarks (UR5Reacher, AntFourRooms, Pendulum)two-level hierarchy · static sparse-reward taskssuccess rateMixed
increase timed subgoal budget (HiTS) (ablation)HiTS · Platformssuccess rateMixed
increase subgoal budget (HAC) (ablation)HAC · Platformssuccess rate: stagnates at around 50%No clear effect
HiTS vs SAC and HAC on Ball in cupHiTS · Ball in cupreturn / speedHelped

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy