Google researchers use deep reinforcement learning for optimizing chip design

Using a trained deep reinforcement learning model, a team at Google calculated the optimal chip placement on an IC—a problem that would have taken several weeks for human experts to solve. Read more…
Neowin