Summary of Self-labeling the Job Shop Scheduling Problem, by Andrea Corsini et al.
Self-Labeling the Job Shop Scheduling Problem
by Andrea Corsini, Angelo Porrello, Simone Calderara, Mauro Dell’Amico
First submitted to arxiv on: 22 Jan 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Combinatorics (math.CO)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A novel self-supervised training strategy is proposed for tackling combinatorial problems, which are notoriously challenging due to the need for costly target solutions produced using exact solvers. The approach draws inspiration from semi- and self-supervised learning, leveraging generative models that can be trained by sampling multiple solutions and using the best one as a pseudo-label. This iterative process enables the model to improve its generation capabilities solely through self-supervision, eliminating the need for optimality information. The proposed Self-Labeling Improvement Method (SLIM) is validated on the Job Shop Scheduling problem, a complex combinatorial issue receiving significant attention from the neural combinatorial community. A generative model based on the Pointer Network is trained using SLIM, and experiments on popular benchmarks demonstrate its potential as the resulting models outperform constructive heuristics and state-of-the-art learning proposals for the JSP. Additionally, the robustness of SLIM to various parameters and its generality are demonstrated by applying it to the Traveling Salesman Problem. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A new way to train computers is introduced that helps solve tricky problems called combinatorial problems. Normally, these problems require a lot of computing power to get the right answers. The new method uses special computer programs called generative models and teaches them to make good guesses by looking at multiple possible solutions. This process keeps improving until the model can generate good solutions on its own without needing a lot of extra help. This approach is tested on a common problem called Job Shop Scheduling and shows that it works better than other methods for this type of problem. |
Keywords
* Artificial intelligence * Attention * Generative model * Self supervised
