Research map / Reinforcement learning / Visual and representation RL
Goal-Conditioned Generators of Deep Policies
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 |
|---|---|---|---|
| Ask for a return 20 above the best return seen so far instead of asking for the current best (ablation) | GoGePo · MuJoCo continuous-control tasks | return | Helped |
| Use the current best return as the command rather than adding a drive parameter of 20 (ablation) | GoGePo · MuJoCo continuous-control tasks | return | No clear effect |
| Scale generator outputs by 2/√n to bring policy weights closer to default initialization (ablation) | GoGePo · Swimmer-v3 | return | Helped |
| Scale generator outputs by 2/√n to bring policy weights closer to default initialization (ablation) | GoGePo · Hopper-v3 | return | Mixed |
| Sample replay-buffer policies with recency-weighted probabilities instead of uniformly (ablation) | GoGePo · Hopper-v3 | return | Helped |
| Use importance-sampling-based command selection instead of the simple command strategy (ablation) | GoGePo · MuJoCo continuous-control tasks | return | No clear effect |
Closest papers by meaning
- Learn Goal-Conditioned Policy with Intrinsic Motivation for Deep Reinforcement Learning (AAAI 2022 · similarity 0.89)
- Flattening Hierarchies with Policy Bootstrapping (NeurIPS 2025 · similarity 0.88)
- $\textit{Hyper-GoalNet}$: Goal-Conditioned Manipulation Policy Learning with HyperNetworks (NeurIPS 2025 · similarity 0.87)
- GFlowNet Training by Policy Gradients (ICML 2024 · similarity 0.87)
- Learning Goal-Conditioned Representations for Language Reward Models (NeurIPS 2024 · similarity 0.86)
- PcLast: Discovering Plannable Continuous Latent States (ICML 2024 · similarity 0.86)
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.