MO-PaDGAN for Design Reparameterization and Optimization

Experiment code associated with our paper on Applied Soft Computing: MO-PaDGAN: Reparameterizing Engineering Designs for Augmented Multi-objective Optimization.

License

This code is licensed under the MIT license. Feel free to use all or portions for your research or related projects so long as you provide the following citation information:

Chen, W., & Ahmed, F. (2021). MO-PaDGAN: Reparameterizing Engineering Designs for augmented multi-objective optimization. Applied Soft Computing, 113, 107909.

@article{chen2021mo,
     title={MO-PaDGAN: Reparameterizing Engineering Designs for augmented multi-objective optimization},
     author={Chen, Wei and Ahmed, Faez},
     journal={Applied Soft Computing},
     volume={113},
     pages={107909},
     year={2021},
     publisher={Elsevier}
   }

Required packages

• tensorflow < 2.0.0
• gpflow
• gpflowopt
• sklearn
• numpy
• matplotlib
• seaborn
• pexpect

Usage

Synthetic examples

1. Go to example directory:

   cd synthetic

2. Train MO-PaDGAN:

   python train.py

   positional arguments:

   mode	train or evaluate
   data	dataset name (specified in datasets.py; available datasets are Ring2D, Grid2D, Donut2D, ThinDonut2D, and Arc2D)
   func	function name (specified in functions.py; available functions are VLMOP2 and NKNO1)
   

   optional arguments:

   -h, --help            	show this help message and exit
   --lambda0		coefficient controlling the weight of quality in the DPP kernel
   --lambda1		coefficient controlling the weight of the performance augmented DPP loss in the PaDGAN loss
   --disc_lr		learning rate for the discriminator
   --gen_lr		learning rate for the generator
   --batch_size		batch size
   --train_steps		training steps
   --save_interval 	number of intervals for saving the trained model and plotting results
   

   The default values of the optional arguments will be read from the file synthetic/config.ini.

   The trained model and the result plots will be saved under the directory synthetic/trained_gan/<data>_<func>/<lambda0>_<lambda1>/, where <data>, <func>, <lambda0>, and <lambda1> are specified in the arguments or in synthetic/config.ini.

   Note that we can set lambda0 and lambda1 to zeros to train a vanilla GAN.

   Datasets and functions are defined in synthetic/functions.py and synthetic/datasets.py, respectively.

   Specifically, here are the dataset and function names (<data> and <func>) for the two synthetic examples in the paper:

   Example	Dataset name	Function name
   I	Ring2D	NKNO1
   II	Ring2D	VLMOP2

3. Multi-objective optimization by Bayesian optimization:

   i) Run single experiment:

   python optimize.py

   positional arguments:

   data	dataset name (specified in datasets.py; available datasets are Ring2D, Grid2D, Donut2D, ThinDonut2D, and Arc2D)
   func	function name (specified in functions.py; available functions are Linear, MixGrid, MixRing, MixRing4, and MixRing6)
   

   optional arguments:

   -h, --help            	show this help message and exit
   --lambda0		coefficient controlling the weight of quality in the DPP kernel
   --lambda1		coefficient controlling the weight of the performance augmented DPP loss in the PaDGAN loss
   --id 	experiment ID
   

   ii) Run experments in batch mode to reproduce results from the paper:

   python run_batch_experiments.py

Airfoil design example

1. Install XFOIL.

2. Go to example directory:

   cd airfoil

3. Download the airfoil dataset here and extract the NPY files into airfoil/data/.

4. Go to the surrogate model directory:

   cd surrogate

5. Train a surrogate model to predict airfoil performances:

   python train_surrogate.py train

   positional arguments:

   mode	train or evaluate
   

   optional arguments:

   -h, --help            	show this help message and exit
   --save_interval		interval for saving checkpoints
   

6. Go back to example directory:

   cd ..

   Train MO-PaDGAN:

   python train.py train

   positional arguments:

   mode	train or evaluate
   

   optional arguments:

   -h, --help            	show this help message and exit
   --lambda0		coefficient controlling the weight of quality in the DPP kernel
   --lambda1		coefficient controlling the weight of the performance augmented DPP loss in the PaDGAN loss
   

   The default values of the optional arguments will be read from the file airfoil/config.ini.

   The trained model and the result plots will be saved under the directory airfoil/trained_gan/<lambda0>_<lambda1>/, where <lambda0> and <lambda1> are specified in the arguments or in airfoil/config.ini. Note that we can set lambda0 and lambda1 to zeros to train a vanilla GAN.

7. Multi-objective optimization by Bayesian optimization:

   i) Run single experiment:

   python optimize_bo.py

   positional arguments:

   parameterization		airfoil parameterization (GAN, SVD, or FFD)
   

   optional arguments:

   -h, --help            	show this help message and exit
   --lambda0		coefficient controlling the weight of quality in the DPP kernel
   --lambda1		coefficient controlling the weight of the performance augmented DPP loss in the PaDGAN loss
   --id 	experiment ID
   

   ii) Run experments in batch mode to reproduce results from the paper:

   python run_batch_experiments_bo.py

8. Multi-objective optimization by Evolutionary Algorithm:

   i) Run single experiment:

   python optimize_ea.py

   positional arguments:

   parameterization		airfoil parameterization (GAN, SVD, or FFD)
   

   optional arguments:

   -h, --help            	show this help message and exit
   --lambda0		coefficient controlling the weight of quality in the DPP kernel
   --lambda1		coefficient controlling the weight of the performance augmented DPP loss in the PaDGAN loss
   --id 	experiment ID
   

   ii) Run experments in batch mode to reproduce results from the paper:

   python run_batch_experiments_ea.py