Hi Marko,

Thanks for your comments.

The code in the tutorial was meant as instructive rather than production-level code. I implemented A3C in a way that I thought would be simple and understandable for a general audience. Running as truly separate processes would indeed provide a speed increase. I am always happy to accept pull requests for any of my tutorials on Github.

In my code each worker is indeed running on a separate core though. Attached is an image of CPU usage during and after running my A3c implementation. As you can see each of the cores is being utilized.

As for the resources I use, I am not sure where I mentioned only using a laptop. For my lab-research I use a dedicated machine with a Titan X Pascal. I also recently began work at a company where we are using cloud GPUs. You are right that anyone with only a laptop would likely be severely limited in the Deep Learning research they could conduct.

PhD. Interests include Deep (Reinforcement) Learning, Computational Neuroscience, and Phenomenology.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store