Parallelize environments and/or training?

jmichaux / intrinsic-motivation

Using multiple sensor modalities to improve exploration for robotic manipulation tasks with sparse rewards

11 stars 6 forks source link

Parallelize environments and/or training? #2

Closed jmichaux closed 5 years ago

jmichaux commented 5 years ago

Right now I am only parallelizing the environments to collect more data using the same agent. Does it make sense to use multiple agents for updating the weights? How would I do this? MPI?