How does binary search apply to finding the peak element in an array?

A peak element is greater than its neighbors. Compare the mid element with mid plus one. If mid is greater, a peak exists on the left side including mid, so set right to mid. If mid plus one is greater, a peak exists on the right side, so set left to mid plus one. This works because if you are going uphill, there must be a peak ahead or at the end. The loop converges when left equals right, pointing to a peak. This finds any peak, not necessarily the global maximum. ---