News for hackers. Stuff that matters.
Re:Spinning Up a Pong AI with Deep Reinforcement Learning by robbiemitchell on Thursday December 06, 2018 @02:12PM (#18621073)
This seems like a solid example of practicing Deep RL with a few lines of code. With the OpenAI thing you run a program from command-line. Curious if you have any insight about what the algorithm is and how it compares to the Keras one?
Re:Spinning Up a Pong AI with Deep Reinforcement Learning by mtrazzi on Thursday December 06, 2018 @02:18PM (#18621135)
Both the Keras one and the one from spinning up can be called "vanilla policy gradients". The one in Keras is closer to REINFORCE, and the one from spinning up use actor-critic and a multilayer perceptron.
Re:Spinning Up a Pong AI with Deep Reinforcement Learning by twillmas on Thursday December 06, 2018 @02:05PM (#18621022)
I’d like to see this AI hooked up to an actual Atari machine somehow. Has anyone tried something like that? Could the model process the frames quickly enough to move a robotic arm up or down on a joystick?
Re:Spinning Up a Pong AI with Deep Reinforcement Learning by mtrazzi on Thursday December 06, 2018 @02:28PM (#18621221)
Hum, have not tried that but I guess the robotic part and the computer vision would be pretty standard, so it would be feasible.

PS:They even have robots playing real Pong! https://www.youtube.com/watch?v=tIIJME8-au8

Re:Spinning Up a Pong AI with Deep Reinforcement Learning by ReDeiPirati on Thursday December 06, 2018 @02:03PM (#18620996)
Great post! You mentioned different RL frameworks in the article (Dopamine, Spinningup, Keras-RL, etc...) Which one do you recommend in term of usability?
Re:Spinning Up a Pong AI with Deep Reinforcement Learning by mtrazzi on Thursday December 06, 2018 @02:14PM (#18621092)
Thanks! I would totally recommend Spinningup, because of the guide and the docs.