Research map / Reinforcement learning / Visual and representation RL
Deep Reinforcement Learning for Scalable Offline Three-Dimensional Packing
Hao Yin, Hongjie He, Fan Chen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add PQNet packed-item encoding (ablation) | OUR (PQNet+BBMP+DCS) · 3D-SPP N=20 (100x100) | space utilization (%): 81.4 -> 82.4 | Improved |
| add PQNet packed-item encoding (ablation) | OUR (PQNet+BBMP+DCS) · 3D-SPP N=50 (100x100) | space utilization (%): 88.4 -> 89.1 | Improved |
| add bidding-based multi-policy (four PQNets) (ablation) | OUR (PQNet+BBMP+DCS) · 3D-SPP N=20 (100x100) | space utilization (%): 81.0 -> 82.4 | Improved |
| add bidding-based multi-policy (four PQNets) (ablation) | OUR (PQNet+BBMP+DCS) · 3D-SPP N=50 (100x100) | space utilization (%): 88.4 -> 89.1 | Improved |
| add DCS instead of candidate queue (ablation) | OUR (PQNet+BBMP+DCS) · 3D-SPP N=100 (100x100) | space utilization (%): 92.3 -> 92.8 | Improved |
| add DCS instead of candidate queue (ablation) | OUR (PQNet+BBMP+DCS) · 3D-SPP N=1000 (100x100) | space utilization (%): 94.5 -> 96.0 | Improved |
| scale to larger containers vs YIN | OUR (PQNet+BBMP+DCS) · 3D-SPP N=50, 400x200 | space utilization (%): 77.1 -> 87.0 | Improved |
| train on N=50 and test on N>=500 (generalization) | OUR (PQNet+BBMP+DCS) · 3D-SPP generalization | space utilization (%) | Hurt |
Closest papers by meaning
- Online 3D Bin Packing with Constrained Deep Reinforcement Learning (AAAI 2021 · similarity 0.89)
- Learning Efficient Online 3D Bin Packing on Packing Configuration Trees (ICLR 2022 · similarity 0.89)
- Distributionally Robust Multi-Agent Reinforcement Learning for Dynamic Chute Mapping (ICML 2025 · similarity 0.82)
- MTL-KD: Multi-Task Learning Via Knowledge Distillation for Generalizable Neural Vehicle Routing Solver (NeurIPS 2025 · similarity 0.82)
- Deep Reinforcement Learning Guided Improvement Heuristic for Job Shop Scheduling (ICLR 2024 · similarity 0.82)
- Graph-Supported Dynamic Algorithm Configuration for Multi-Objective Combinatorial Optimization (ICML 2025 · similarity 0.82)
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.
