Batched Data-Driven Evolutionary Multi-Objective Optimization Based on Manifold Interpolation

Batched Data-Driven Evolutionary Multi-Objective Optimization Based on Manifold Interpolation Ke Li*, Renzhi Chen* [Paper] [Supplementary]

Overview

This repository contains Python implementation of the algorithm framework for Batched Data-Driven Evolutionary Multi-Objective Optimization Based on Manifold Interpolation.

Code Structure

algorithms/ --- algorithms definitions problems/ --- multi-objective problem definitions revision/ -- patch for Gpy package scripts/ --- scripts for batch experiments ├── build.sh --- complie the c lib for test problems ├── run.sh -- run the experiment main.py --- main execution file

Requirements

• Python version: tested in Python 3.7.7
• Operating system: tested in Ubuntu 20.04

Getting Started

Basic usage

Run the main file with python with specified arguments:

python3.7 main.py --problem dtlz7 --n-var 6 --n-obj 2

Parallel experiment

Run the script file with bash, for example:

./scripts/run.sh

Result

The optimization results are saved in txt format. They are stored under the folder:

output/data/{problem}/x{n}y{m}/{algo}-{exp-name}/{seed}/

Citation

If you find our repository helpful to your research, please cite our paper:

@article{KeLi2022,
  title={Batched Data-Driven Evolutionary Multi-Objective Optimization Based on Manifold Interpolation},
  author={Li, Ke and Chen, Renzhi},
  journal={IEEE Transactions on Evolutionary Computation},
  year={2022},
  publisher={IEEE}
}