The problem
Find any peak: a position whose value is bigger than both neighbours (outside the list counts as −∞).
Input: [1, 2, 1, 3, 5, 6, 4] Output: 5 (value 6; position 1 is also a peak)
Picture it
Picture a hiker in fog who can only see neighbouring spots. If the trail goes uphill to the right, keep walking right: you must hit a summit eventually. Binary search does this by checking the middle and moving towards the uphill side.
Short version: the full lesson (brute force, complexity, edge cases and a quiz) is still being written. The animation is complete.
The key idea
If nums[mid] < nums[mid + 1], a peak lies to the right; otherwise to the left (or at mid). O(log n).
Python solution
def find_peak(h):
lo, hi = 0, len(h) - 1
while lo < hi:
mid = (lo + hi) // 2
if h[mid] < h[mid + 1]:
lo = mid + 1 # uphill to the right
else:
hi = mid # a peak is at mid or left
return loIn the interactive version each line lights up as the animation runs, and you can type your own input to see the algorithm handle it.