Four short links: 18 May 2016

The Atari Domain, Designing Experiments, AI Challenge, and Future of Coding

By Nat Torkington
May 18, 2016
Four short links
  1. Asynchronous Methods for Deep Reinforcement LearningThe best performing method, an asynchronous variant of actor-critic, surpasses the current state-of-the-art on the Atari domain while training for half the time on a single multi-core CPU instead of a GPU. Furthermore, we show that asynchronous actor-critic succeeds on a wide variety of continuous motor control problems as well as on a new task involving finding rewards in random 3D mazes using a visual input. Two things: “on the Atari domain” (12-year-old me would love to know this is the state of the art in AI measurement) and “actor-critic” (12-year-old me would be very familiar with the idea of an internal critic).
  2. A First Course in Design and Analysis of Experiments (PDF) — textbook on designing experiments. Golden.
  3. Learn faster. Dig deeper. See farther.

    Join the O'Reilly online learning platform. Get a free trial today and find answers on the fly, or master something new and useful.

    Learn more
  4. Vindiniuman Artificial Intelligence programming challenge. You have to take the control of a legendary hero using the programming language of your choice. You will fight with other AI for a predetermined number of turns, and the hero with the greatest amount of gold will win. See earlier comment about 12-year-old me.
  5. The End of Code (Wired) — “I was just having a conversation about that this morning,” says tech guru Tim O’Reilly when I ask him about this shift. “I was pointing out how different programming jobs would be by the time all these STEM-educated kids grow up.” Traditional coding won’t disappear completely—indeed, O’Reilly predicts that we’ll still need coders for a long time yet—but there will likely be less of it, and it will become a meta skill, a way of creating what Oren Etzioni, CEO of the Allen Institute for Artificial Intelligence, calls the “scaffolding” within which machine learning can operate. Future of coding (AI? formal methods? provably secure?) is super interesting to me. Send me links if there’s something you see that interests you: @gnat on Twitter. kthx.
Post topics: Four Short Links