As mentioned earlier, the boundaries defined in the learned model are not perfect. Some unlabeled snippets will land in the wrong regions and, as a result, step count errors will be made. For example, during the course of the development of the classifier we have launched with UP2 and UP3 we encountered a number of situations where errors were made. Our VP of Software noticed one day that his steps were undercounted as he walked back to his desk carefully holding a full cup of coffee – snippets were landing in the wrong classifier region. In order to address this problem we needed to adjust the region boundaries and, for this, we needed to provide the machine learning algorithm with additional examples – examples of the problematic behavior to be precise.

When we first began to explore the idea of building a step classifier, we knew we would be constrained to a very limited population of individuals (Jawbone employees) available to us for early development and testing. It seemed certain that the development of the classifier would be very iterative in that, as we tested larger and more varied sets of individuals and behaviors, we would undoubtedly find issues that we needed to quickly correct. So we would need a technical approach that was suited to rapid updates and that those updates would need to be essentially risk free. We could not afford the risk and development time of actually writing new code as we iterated. In short, we needed a step classifier that learned.