Research map / Reinforcement learning / Visual and representation RL
The Laplacian in RL: Learning Representations with Efficient Approximations
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 the graph drawing objective to learn Laplacian representations | State representation learning · FourRoom | graph drawing objective gap | Helped |
| Use Laplacian-distance reward mixed with sparse reward | DQN · TwoRooms and HardMaze | learning performance | Helped |
| Use Laplacian-distance reward mixed with sparse reward | DQN · OneRoom | learning performance | Mixed |
| Use Laplacian-distance reward shaping, embedding goal coordinates | DDPG · Continuous-control navigation | learning performance | Helped |
| Use Laplacian-distance reward shaping, embedding the full state | DDPG · Continuous-control navigation | learning performance | Helped |
| Mix the learned-distance reward with sparse reward instead of using learned distance alone (ablation) | Laplacian reward shaping · GridWorld | learning performance near the goal | Helped |
| Vary the orthonormality penalty weight β (ablation) | Graph drawing representation learning · FourRoom, position representation | graph drawing objective gap | Mixed |
| Learn representations online instead of pretraining and fixing them (ablation) | Laplacian reward shaping · Continuous-control navigation | learning performance | No clear effect |
Closest papers by meaning
- Towards Better Laplacian Representation in Reinforcement Learning with Generalized Graph Drawing (ICML 2021 · similarity 0.94)
- Online Laplacian-Based Representation Learning in Reinforcement Learning (ICML 2025 · similarity 0.94)
- 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 0.89)
- BeigeMaps: Behavioral Eigenmaps for Reinforcement Learning from Images (ICML 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.