Research map / Reinforcement learning / Visual and representation RL
Towards Better Laplacian Representation in Reinforcement Learning with Generalized Graph Drawing
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 |
|---|---|---|---|
| Replace the unweighted spectral graph drawing objective with the generalized objective using strictly decreasing coefficients | Laplacian representation learning · GridRoom, GridMaze, GridRoom (image), GridMaze (image), PointRoom, PointMaze | SimGT: GridRoom: 0.239 -> 0.991; GridMaze: 0.220 -> 0.962; GridRoom (image): 0.310 -> 0.985; GridMaze (image): 0.229 -> 0.984; | Helped |
| Use Group 1 coefficients instead of the default coefficient schedule (ablation) | Generalized graph drawing · GridRoom | similarity to ground truth: default 0.9913; group 1 0.9905 | No clear effect |
| Use Group 2 coefficients instead of the default coefficient schedule (ablation) | Generalized graph drawing · GridRoom | similarity to ground truth: default 0.9913; group 2 0.9653 | No clear effect |
| Use the learned generalized-objective representation rather than the baseline representation to discover options | Option discovery · GridRoom and GridMaze | option exploration across time scales | Helped |
| Use the learned generalized-objective representation rather than the baseline representation for reward shaping | Reward shaping · GridRoom and GridMaze | goal-reaching success rate | Helped |
| Use a lower representation dimension instead of all dimensions for reward shaping (ablation) | Reward shaping with the learned representation · GridRoom and GridMaze | goal-reaching success rate | Helped |
Closest papers by meaning
- Online Laplacian-Based Representation Learning in Reinforcement Learning (ICML 2025 · similarity 0.95)
- Proper Laplacian Representation Learning (ICLR 2024 · similarity 0.94)
- The Laplacian Keyboard: Beyond the Linear Span (NeurIPS 2026 · similarity 0.91)
- Diffusion Spectral Representation for Reinforcement Learning (NeurIPS 2024 · similarity 0.87)
- BeigeMaps: Behavioral Eigenmaps for Reinforcement Learning from Images (ICML 2024 · similarity 0.86)
- Geometry of Neural Reinforcement Learning in Continuous State and Action Spaces (ICLR 2025 · 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.