The Pareto-Weighted-Sum-Tuning (PWST) utilizes Learning-to-Rank Machine Learning to help solve Pareto (Multiobjective, Multicriteria) Optimization Problems. It was proposed and invented by Harry Wang during his time as an Undergraduate Researcher and Computer Science student at the University of Michigan-Ann Arbor, College of Engineering. Harry Wang is currently a Computer Science Graduate Student at the University of Pennsylvania, School of Engineering and Applied Science, where he focuses in Machine Learning and Natural Language Processing. This codebase utilizes PWST to run experiments with a sample user/decision-maker on an example stock-pricing dataset.
Pareto-Weighted-Sum-Tuning was accepted and presented at the 2020 International Conference on Machine Learning, Computational Optimization, and Data Science (LOD) (https://lod2020.icas.xyz/program/).
LOD 2020 Slides: https://docs.google.com/presentation/d/1WGXAiDIJQ18kT1rCZL_hl6W0rJk9UgFT4g3nL_Fu-AM/edit?usp=sharing
Publication Link: https://www.springerprofessional.de/en/pareto-weighted-sum-tuning-learning-to-rank-for-pareto-optimizat/18742156
The weighted-sum method is a commonly used technique in Multi-objective optimization to represent different criteria considered in a decision-making and optimization problem. Weights are assigned to different criteria depending on the degree of importance. However, even if decision-makers have an intuitive sense of how important each criteria is, explicitly quantifying and hand-tuning these weights can be difficult. To address this problem, we propose the Pareto-Weighted-Sum-Tuning algorithm as an automated and systematic way of trading-off between different criteria in the weight-tuning process. Pareto-Weighted-Sum-Tuning is a configurable online-learning algorithm that uses sequential discrete choices by a decision-maker on sequential decisions, eliminating the need to score items or weights. We prove that utilizing our online-learning approach is computationally less expensive than batch-learning, where all the data is available in advance. Our experiments show that Pareto-Weighted-Sum-Tuning is able to achieve low relative error with different configurations.
To run a sample PWST experiment referenced, run the following command. Graphs will be generated to display the results produced by PWST.
python example.py