Research map / Reinforcement learning / Policy gradient and actor-critic
Distributional Pareto-Optimal Multi-Objective 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 DPMORL instead of GPI-PD | GPI-PD · Hopper | Expected Utility: 2374.44 -> 3492.93 | Helped |
| Use DPMORL instead of GPI-PD | GPI-PD · HalfCheetah | Hypervolume: 1083227.57 -> 8593769.62 | Helped |
| Use DPMORL instead of GPI-PD | GPI-PD · DeepSeaTreasure | Expected Utility: 5.93 -> 6.70 | Helped |
| Use DPMORL instead of GPI-PD | GPI-PD · DeepSeaTreasure | Hypervolume: 9.75 -> 8.68 | Hurt |
| Use DPMORL instead of GPI-PD | GPI-PD · FruitTree | Expected Utility: 6.98 -> 6.89 | No clear effect |
| Use DPMORL instead of GPI-PD | GPI-PD · Hopper | Constraint Satisfaction: 0.47 -> 0.76 | Helped |
| Use DPMORL instead of GPI-PD | GPI-PD · DeepSeaTreasure | Variance Objective: 2.59 -> 2.54 | No clear effect |
| Use DPMORL instead of GPI-PD | GPI-PD · FruitTree | Variance Objective: 3.42 -> 3.21 | Hurt |
| Use DPMORL instead of GPI-PD with 3 objectives | GPI-PD · Hopper 3-objective | Expected Utility: 1511.40 -> 2107.26 | Helped |
| Increase number of policies from 5 to 10 (ablation) | DPMORL · Hopper | Expected Utility: 3274.36 -> 3309.52 | Helped |
| Increase number of policies from 5 to 15 (ablation) | DPMORL · Hopper | Hypervolume: 9917663.14 -> 11453079.82 | Helped |
| Increase number of policies from 5 to 10 (ablation) | DPMORL · DeepSeaTreasure | Hypervolume: 8.93 -> 14.88 | Helped |
Closest papers by meaning
- PA2D-MORL: Pareto Ascent Directional Decomposition Based Multi-Objective Reinforcement Learning (AAAI 2024 · similarity 0.91)
- Pareto Set Learning for Multi-Objective Reinforcement Learning (AAAI 2025 · similarity 0.91)
- Preference Conditioned Multi-Objective Reinforcement Learning: Decomposed, Diversity-Driven Policy Optimization (NeurIPS 2026 · similarity 0.89)
- Preference Optimization on Pareto Sets: On a Theory of Multi-Objective Optimization (NeurIPS 2025 · similarity 0.89)
- On Generalization Across Environments In Multi-Objective Reinforcement Learning (ICLR 2025 · similarity 0.89)
- An Analytical Study of Utility Functions in Multi-Objective Reinforcement Learning (NeurIPS 2024 · similarity 0.89)
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.