Katkat
Back

Compute moving average over last N stream

AtlassianMachine Learning Engineer · Onsite
MediumMachine Learning EngineerOnsiteCoding & Algorithms

Last updated

This question evaluates understanding of streaming algorithms and sliding-window aggregation within the Coding & Algorithms domain, focusing on incremental computation, state management, and numerical stability when computing moving averages.

You are given an integer N and an unbounded stream of integers arriving one by one. After each new integer arrives, output the average of the last N integers seen so far.

Requirements:

  • If fewer than N integers have arrived, average over all integers seen so far.
  • The solution should support efficient updates per element.

Follow-up:

  • How would you modify the design if more recent numbers should have higher weight (e.g., exponentially decaying weights or a custom weight vector over the last N items)?