Hands-On Neuroevolution with Python

Setting up the work environment

The work environment for training the agent to play Atari games assumes that a large controller ANN needs to be trained in the process. We already stated that the controller ANN has more than 4 million training parameters and requires a lot of computational resources to be able to evaluate. Fortunately, modern GPU accelerators allow the execution of massive parallel computations simultaneously. This feature is convenient for our experiment because we need to evaluate each individual against the game environment multiple times during the evolution process. Without GPU acceleration, it would either take a lot of time or require a massive number of processing cores (about 720).

Let's discuss how to prepare the ...