Confidence Threshold Neural Diving
This work addresses the challenge of improving solution quality and speed in optimization for researchers and practitioners, though it is incremental as it builds on existing Neural Diving methods.
The paper tackles the problem of finding better feasible solutions faster for Mixed Integer Programs by proposing a confidence threshold technique within Neural Diving, which achieved 2nd place in the NeurIPS 2021 ML4CO competition's primal task and outperformed other learning-based methods.
Finding a better feasible solution in a shorter time is an integral part of solving Mixed Integer Programs. We present a post-hoc method based on Neural Diving to build heuristics more flexibly. We hypothesize that variables with higher confidence scores are more definite to be included in the optimal solution. For our hypothesis, we provide empirical evidence that confidence threshold technique produces partial solutions leading to final solutions with better primal objective values. Our method won 2nd place in the primal task on the NeurIPS 2021 ML4CO competition. Also, our method shows the best score among other learning-based methods in the competition.