Research map / Reinforcement learning / Visual and representation RL
Learning Object Relation Graph and Tentative Policy for Visual Navigation
Heming Du, Xin Yu, Liang Zheng
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add trial-driven IL (ablation) | navigation network · AI2-Thor unseen scenes (ALL) | success rate (%): 66.8% -> 69.3% | Improved |
| add trial-driven IL (ablation) | navigation network · AI2-Thor unseen scenes (ALL) | SPL: 0.375 -> 0.394 | Improved |
| add ORG (ablation) | navigation network · AI2-Thor unseen scenes (ALL) | success rate (%): 67.5% -> 69.3% | Improved |
| add ORG (ablation) | navigation network · AI2-Thor unseen scenes (ALL) | SPL: 0.345 -> 0.394 | Improved |
| add TPN (ablation) | navigation network · AI2-Thor unseen scenes (ALL) | success rate (%): 66.6% -> 69.3% | Improved |
| add TPN (ablation) | navigation network · AI2-Thor unseen scenes (ALL) | SPL: 0.387 -> 0.394 | Improved |
| apply IL supervision at every step instead of only in deadlock (ablation) | navigation network · AI2-Thor unseen scenes (ALL) | success rate (%): 69.3% -> 47.7% | Hurt |
| update navigation network via TPN at every state (ablation) | navigation network · AI2-Thor unseen scenes (ALL) | success rate (%): 69.3% -> 66.2% | Hurt |
| replace TPN escape policy with random actions (ablation) | navigation network · AI2-Thor unseen scenes (ALL) | success rate (%): 69.3% -> 62.3% | Hurt |
| add ORG to baseline | baseline · AI2-Thor unseen scenes (ALL) | success rate (%): 56.4 -> 65.3 | Improved |
| add trial-driven IL to baseline | baseline · AI2-Thor unseen scenes (ALL) | success rate (%): 56.4 -> 63.6 | Improved |
| add TPN to baseline | baseline · AI2-Thor unseen scenes (ALL) | SPL: 0.319 -> 0.316 | No clear effect |
Closest papers by meaning
- VTNet: Visual Transformer Network for Object Goal Navigation (ICLR 2021 · similarity 0.89)
- Object-Goal Visual Navigation via Effective Exploration of Relations Among Historical Navigation States (CVPR 2023 · similarity 0.88)
- DMC-VB: A Benchmark for Representation Learning for Control with Visual Distractors (NeurIPS 2024 · similarity 0.86)
- Robust Visual Imitation Learning with Inverse Dynamics Representations (AAAI 2024 · similarity 0.86)
- What's the Move? Hybrid Imitation Learning via Salient Points (ICLR 2025 · similarity 0.85)
- When Should We Prefer State-to-Visual DAgger over Visual Reinforcement Learning? (AAAI 2025 · similarity 0.85)
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.
