search

hzwer / ICCV2019-LearningToPaint

ICCV2019 - Learning to Paint With Model-based Deep Reinforcement Learning
MIT License
2.25k stars 312 forks source link

Role of constant coordinates in merged state? #37

Closed 1jsingh closed 4 years ago

1jsingh commented 4 years ago

Hi,

Can you please clarify why the merged state for actor, critic takes the constant coordinate images as input?

Thanks!

hzwer commented 4 years ago

The parameters of the convolutional neural network are position-independent, while our action parameter space is coordinate-dependent. In our experiment, adding coordinate features can slightly improve the network performance.