using GAN in regression proplems

Greetings,

I am doing a research paper to invetigate the capabilities of Deep Learning in predicting Residual Stress in metals after welding. however, there is very little information from deep learning experts tothis end. one of my ideas is using GAN to generate more data, as data of RS post welding is very scarce. however, all over the internet and in matlab especially 99% of Deep Learning is in image recognition. anyhow, is there any ideas of how to use GAN to ptoduce numeric tabular results?

Thank you and sorry for the very lengthy introduction.

in anyway if anyone has any advice disregarding my question it would be very much appreciated.

Answer 1

[Abhishek Gupta]

Hi,

As per my understanding, you want to use a GAN network to generate numeric tabular data for your research work. One can achieve this task by using fullyConnectedLayer in place of convolution2dLayer or transposedConv2dLayer in the generator and the discriminator model.

Further, you can refer to the following link to train your GAN model: -

https://www.mathworks.com/help/deeplearning/ug/train-generative-adversarial-network.html

[Abdulaziz Almershed]

thank you very much