Research map / Reinforcement learning / Visual and representation RL
Goal Recognition as Reinforcement 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 |
|---|---|---|---|
| Use learned Q-functions with MaxUtil instead of planning-based R&G | GRAQL · Blocks, partial observability (Avg) | accuracy: 0.78 -> 0.98 | Helped |
| Use learned Q-functions with KL instead of planning-based R&G | GRAQL · Hanoi, partial observability (Avg) | accuracy: 0.68 -> 0.95 | Helped |
| Use KL rather than planning-based R&G | GRAQL · SkGrid, partial observability (Avg) | accuracy: 0.86 -> 0.94 | Helped |
| Use MaxUtil rather than KL for noisy observations | GRAQL · Blocks, noisy observations (Avg) | accuracy: 0.81 -> 0.97 | Helped |
| Use MaxUtil rather than KL for noisy observations | GRAQL · Hanoi, noisy observations (Avg) | accuracy: 0.93 -> 0.99 | Helped |
| Use KL rather than MaxUtil for noisy observations | GRAQL · SkGrid, noisy observations (Avg) | accuracy: 0.80 -> 0.85 | Helped |
| Initialize Q-functions from one planner-generated optimal trajectory (policy shaping) (ablation) | GRAQL · Blocks, Hanoi, and SkGrid | recognition performance | No clear effect |
| Train for 500 rather than 10k or 30k episodes (ablation) | GRAQL · Blocks, Hanoi, and SkGrid | greedy policy plan length: around 5% | No clear effect |
| Use action-only observations rather than state-action observations (ablation) | MaxUtil · SkGrid | accuracy: 0.54 | Hurt |
| Use action-only observations rather than state-action observations (ablation) | MaxUtil · SkGrid | precision: 0.23 | Hurt |
Closest papers by meaning
- Autonomous Goal Detection and Cessation in Reinforcement Learning: A Case Study on Source Term Estimation (AAAI 2025 · similarity 0.87)
- From Weak Data to Strong Policy: Q-Targets Enable Provable In-Context Reinforcement Learning (NeurIPS 2026 · similarity 0.87)
- Offline Goal-conditioned Reinforcement Learning with Quasimetric Representations (NeurIPS 2025 · similarity 0.87)
- Inverting the Bellman Equation: From $Q$-Values to World Models (NeurIPS 2026 · similarity 0.86)
- Pre-Trained Multi-Goal Transformers with Prompt Optimization for Efficient Online Adaptation (NeurIPS 2024 · similarity 0.86)
- Learning to Reach Goals via Diffusion (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.