What I love about covering science: I did a deep dive - OK, maybe just a shallow dive - into artificial intelligence just to be able to write a coherent one-sentence explanation of deep reinforcement learning. Working with old friends
at UC Berkeley's College of Engineering and Institute of Transportation Studies on this story about incorporating vehicle automation into traffic-management strategies, I faced a formidable explanation task, pulling back the curtain on a notable development in traffic-management technology and why it is - or soon will be - important to all of us. The story yielded this great coverage in Science Magazine and this spot on the local ABC News affiliate.
The study specifies highly detailed scenarios — standard “tasks” that engineers can use to solve common types of traffic challenges like bottlenecks and intersection control. The solutions become shared baselines, called benchmarks, that are critical to making progress, researchers say. “Unless we’re working on the same problem, it’s hard to compare results. Are you looking at a New York highway or a California freeway? A group of 20 cars or 50? You need an apples-to-apples comparison to understand which solution works better,” Vinitsky said. Read more.