Research map / Learning theory and algorithms / Combinatorial and search optimization
From Sequential to Recursive: Enhancing Decision-Focused Learning with Bidirectional Feedback
Xinyu Wang, Jinxiao Du, Yiyang Peng, Wei Ma
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 decision feedback to the predictor (recursive R-DFL-I instead of sequential S-DFL) | DFL · Newsvendor (small, n=10) | RMSE: 12.245 -> 8.831 | Improved |
| add decision feedback to the predictor (recursive R-DFL-I instead of sequential S-DFL) | DFL · Newsvendor (large, n=100) | RMSE: 12.649 -> 9.327 | Improved |
| add decision feedback (R-DFL-U instead of S-DFL) | DFL · Newsvendor (mid, n=50) | RMSE: 12.536 -> 9.173 | Improved |
| add decision feedback (R-DFL-I instead of S-DFL) | DFL · Bipartite matching (large, n=900) | RMSE: 0.187 -> 0.171 | Improved |
| add decision feedback (R-DFL-U instead of S-DFL) | DFL · Bipartite matching (small, n=16) | RMSE: 0.408 -> 0.396 | Improved |
| add decision feedback (R-DFL-I instead of S-DFL) | DFL with Transformer predictor · Newsvendor (large) | RMSE: 14.040 -> 11.332 | Improved |
| add decision feedback (R-DFL-I instead of S-DFL) | DFL with Transformer predictor · Bipartite matching (large) | RMSE: 0.188 -> 0.166 | Improved |
| add decision feedback (R-DFL-I instead of S-DFL) | DFL with LSTM predictor · Newsvendor (large) | RMSE: 12.693 -> 10.104 | Improved |
| use implicit differentiation instead of explicit unrolling | R-DFL · Bipartite matching (large) | training time per epoch (s): 2704 -> 1867 | Improved |
| use implicit differentiation instead of explicit unrolling | R-DFL · Bipartite matching (mid) | training time per epoch (s): 432 -> 65 | Improved |
| use implicit differentiation instead of explicit unrolling | R-DFL · Newsvendor (large, n=100) | RMSE: 9.343 -> 9.327 | No clear effect |
| use implicit differentiation instead of explicit unrolling | R-DFL · Bipartite matching (small, n=16) | RMSE: 0.396 -> 0.398 | Hurt |
Closest papers by meaning
- Solver-Free Decision-Focused Learning for Linear Optimization Problems (NeurIPS 2025 · similarity 0.88)
- DFF: Decision-Focused Fine-Tuning for Smarter Predict-Then-Optimize with Limited Data (AAAI 2025 · similarity 0.88)
- Decision-Focused Learning with Directional Gradients (NeurIPS 2024 · similarity 0.87)
- Feasibility-Aware Decision-Focused Learning for Predicting Parameters in the Constraints (NeurIPS 2025 · similarity 0.86)
- Locally Convex Global Loss Network for Decision-Focused Learning (AAAI 2025 · similarity 0.85)
- In-Context Decision Transformer: Reinforcement Learning via Hierarchical Chain-of-Thought (ICML 2024 · similarity 0.85)
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.
