b'SELECTING OUR CITIESOur multicity testingprocess is focused on balancing diversity of location with rigorous testing to deeply understand the details in each. Our intent is to dig deep into the road behaviors in each city we select. Hence, weve striven to identify the set of cities that would expose our vehicles to the widest possible variety of interactions and behaviors. Wealsotestinmultiplecitiesinordertoavoidoverfitting,acommonma-chine-learning problem when what works well in one environment does not translate to another. (This is a bit like a retailer using a one-size-fits-all strategy for selling cloth-ing, and we all have seen how that turns out.) Instead, were taking a more tailored approach to developing our self-driving system. We dont want to create a self-driving vehicle that operates well in Miami traffic but cant adapt to other locations.The system also needs to understand that traffic infrastructure can vary widely. We want our SDS to be able to identify and handle the vertical traffic lights in one city as well as the horizontal ones in another. To avoid overfitting, we carefully select our test cities to provide our system with a broad array of challenges that are representative of what it might encounter in cities around the world. Understanding all the variables in different driving environments is difficult. But testing in multiple cities now will ultimately enable us to expand into others faster, reducing the time it will take for our partners to deploy services across the globe. While the initial commercialization of self-driving technology may be on a street-by-street, block-by-block basis, the rubber will really meet the road with our ability to expandThe Argo SDS is designed city-to-city.to handle a diversity of A development approach centered on dense urban centers delivers the complexitycomplexity, whether test-ing in Washington, D.C., required to prepare our self-driving system for the real worldbecause thats the world(top) or in Pittsburgh.we live in.FullBook_Mar24.indb 119 4/25/21 6:44 PM'