Research map / Reinforcement learning / Policy gradient and actor-critic
Adjustable Robust Reinforcement Learning for Online 3D Bin Packing
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 |
|---|---|---|---|
| Balance nominal and adversarial returns using exact AR2L | Online 3D-BPP packing policy · Mixed-Item Dataset, NB=20 | Uti.: 48.7 -> 52.9 | Helped |
| Use exact AR2L instead of RARL | Online 3D-BPP packing policy · Discrete 3D-BPP, NB=5, β=100 | Uti.: 66.7 -> 66.5 | No clear effect |
| Increase robustness weight α from 0.3 to 1.0 (ablation) | ExactAR2L · Discrete 3D-BPP, NB=5 | Uti.: 76.3 -> 77.4; 64.5 -> 66.5 | Mixed |
| Increase robustness weight α from 0.3 to 1.0 (ablation) | ExactAR2L · Continuous 3D-BPP, NB=5, β=0 | Uti.: 65.9 -> 66.7 | Helped |
| Use approximate value estimation instead of mixture-dynamics samples (ablation) | AR2L · Continuous 3D-BPP, NB=5, β=0 | Uti.: 67.0 -> 67.4 | Helped |
| Use approximate value estimation instead of mixture-dynamics samples (ablation) | AR2L · Discrete 3D-BPP | Uti. | Hurt |
| Increase attacker observation window NB | Permutation-based attacker · Discrete 3D-BPP, all evaluated methods | Uti. | Hurt |
Closest papers by meaning
- Breaking the Barrier: Enhanced Utility and Robustness in Smoothed DRL Agents (ICML 2024 · similarity 0.86)
- Adversarial Diffusion for Robust Reinforcement Learning (NeurIPS 2025 · similarity 0.85)
- Time-Constrained Robust MDPs (NeurIPS 2024 · similarity 0.85)
- Distributionally Robust Multi-Agent Reinforcement Learning for Dynamic Chute Mapping (ICML 2025 · similarity 0.85)
- Towards Robust Offline Reinforcement Learning under Diverse Data Corruption (ICLR 2024 · similarity 0.84)
- Towards Robust Model-Based Reinforcement Learning Against Adversarial Corruption (ICML 2024 · similarity 0.84)
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.