Research map / Reinforcement learning / Imitation learning
Goal-conditioned Imitation Learning
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 changed | Where | Measured | Effect |
|---|---|---|---|
| combine hindsight experience replay with goal-conditioned GAIL and demonstrations | goal-conditioned policy learning · Four rooms, Pointmass Block Pusher, Fetch Pick & Place, Fetch Stack Two | coverage | Helped |
| add expert relabeling by assigning demonstrated transitions future states as goals (ablation) | Behavioral Cloning, BC+HER, and goalGAIL · Four rooms, Pointmass Block Pusher, Fetch Pick & Place, Fetch Stack Two | coverage | Helped |
| combine behavioral cloning with HER | goal-conditioned policy learning · Four rooms, Pointmass Block Pusher, Fetch Pick & Place, Fetch Stack Two | coverage | Helped |
| anneal the behavioral cloning loss weight (ablation) | BC+HER · goal-conditioned tasks | coverage | Helped |
| use goal-conditioned GAIL without combining it with HER (ablation) | goal-conditioned policy learning · Four rooms, Pointmass Block Pusher, Fetch Pick & Place, Fetch Stack Two | coverage | Hurt |
| use HER without demonstrations (ablation) | goal-conditioned policy learning · Four rooms, Pointmass Block Pusher, Fetch Pick & Place, Fetch Stack Two | coverage | Hurt |
| replace expert actions with state transitions in the discriminator (ablation) | goalGAIL · Four rooms, Pointmass Block Pusher, Fetch Stack Two | coverage | Helped |
| replace expert actions with state transitions in the discriminator (ablation) | goalGAIL · Fetch Pick & Place | coverage | Mixed |
| train from noisy, sub-optimal expert demonstrations using a discriminator instead of directly copying actions (ablation) | goalGAIL versus Behavioral Cloning · Four rooms, Pointmass Block Pusher, Fetch Pick & Place, Fetch Stack Two | coverage | Helped |
| use noisy sub-optimal demonstrations with Behavioral Cloning (ablation) | Behavioral Cloning · Four rooms, Pointmass Block Pusher, Fetch Pick & Place, Fetch Stack Two | coverage | Hurt |
| use state-only demonstrations instead of state-action demonstrations (ablation) | goalGAIL · goal-conditioned tasks | coverage | Mixed |
| add HER to Behavioral Cloning (ablation) | Behavioral Cloning · goal-conditioned tasks | coverage | Helped |
Closest papers by meaning
- Goal-Conditioned On-Policy Reinforcement Learning (NeurIPS 2024 · similarity 0.90)
- Imitation Learning from Observation with Automatic Discount Scheduling (ICLR 2024 · similarity 0.89)
- Learning from Demonstrations via Capability-Aware Goal Sampling (NeurIPS 2025 · similarity 0.89)
- Expert Proximity as Surrogate Rewards for Single Demonstration Imitation Learning (ICML 2024 · similarity 0.89)
- Cluster-Based Sampling in Hindsight Experience Replay for Robotic Tasks (Student Abstract) (AAAI 2024 · similarity 0.88)
- GO-DICE: Goal-Conditioned Option-Aware Offline Imitation Learning via Stationary Distribution Correction Estimation (AAAI 2024 · similarity 0.88)
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.