Fast and Interpretable Mixed-Integer Linear Program Solving by Learning Model Reduction
Authors: Yixuan Li, Can Chen, Jiajun Li, Jiahui Duan, Xiongwei Han, Tao Zhong, Vincent Chau, Weiwei Wu, Wanyuan Wang
AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Evaluation on real-world MILP problems shows that 1) compared to the state-of-the-art model reduction ML methods, our method obtains nearly 20% improvement on solution accuracy, and 2) compared to the commercial solver Gurobi, two to four orders of magnitude speedups are achieved. In this section, we compare our proposed method with the learning-based methods and the commercial solvers on real-world datasets to validate the performance and efficiency. |
| Researcher Affiliation | Collaboration | 1School of Computer Science and Engineering, Southeast University 2Noah s Ark Lab, Huawei Technologies EMAIL, EMAIL |
| Pseudocode | Yes | Algorithm 1: Preference-based Strategy Learning Input: Training data {[θi, SP ] = [ θi, s1 , ..., θi, s M P ] | θi ΘN}, rewards r = {r(θi, sj) | θi ΘN, sj SP }, preference model Rϕ, learning rate α, weight λ1 and λ2. Output: Optimized model parameters Rϕ. 1: Initialize model parameters Rϕ. 2: for each instance θi do ... |
| Open Source Code | No | The paper does not provide any statement or link regarding the public availability of source code for the methodology described. |
| Open Datasets | Yes | 1. MIPLIB (Gleixner et al. 2021), six scenarios selected as in (Bertsimas and Kim 2023), the real-world MILP problems with varying scales and solving difficulties. 2. Fuel Cell Energy Management Problem (Frick, Domahidi, and Morari 2015), treated as the primary evaluation scenario by (Bertsimas and Stellato 2022). |
| Dataset Splits | Yes | For Datasets 1 and 3, we used 1000 samples for training and 500 for testing; for Dataset 2, 10,000 samples were used for training and 1000 for testing. |
| Hardware Specification | No | The paper mentions comparing against Gurobi, a commercial solver, but does not provide specific details about the hardware (e.g., GPU/CPU models) used for running its own experiments. |
| Software Dependencies | No | The paper mentions using commercial solvers such as 'Gurobi, Cplex and Matlab' but does not provide specific version numbers for these or any other software libraries or frameworks used in the implementation. |
| Experiment Setup | Yes | where the tolerances ϵ1 for infeasibility and ϵ2 for suboptimality are both set to 1 10 4. The total loss is: Ltotal(ϕ) = λ1Lp(ϕ) + λ2Ld(ϕ), where λ1 and λ2 are hyperparameters for different scenarios. Algorithm 1 also lists 'learning rate α, weight λ1 and λ2' as inputs. |