Formalizing Interruptible Algorithms for Human over-the-loop Analytics

TL;DR

This paper proposes a formal framework for interruptible human-in-the-loop algorithms in data analytics, aiming to balance accuracy improvements with minimal increases in execution time by allowing user intervention only when necessary.

Contribution

It introduces a formalization of interruptible algorithms that enable human intervention over-the-loop without requiring constant user feedback, enhancing efficiency and accuracy.

Findings

Framework for interruptible algorithms formalized

Potential for improved accuracy with minimal time loss

Addresses challenges of human-in-the-loop data mining

Abstract

Traditional data mining algorithms are exceptional at seeing patterns in data that humans cannot, but are often confused by details that are obvious to the organic eye. Algorithms that include humans "in-the-loop" have proved beneficial for accuracy by allowing a user to provide direction in these situations, but the slowness of human interactions causes execution times to increase exponentially. Thus, we seek to formalize frameworks that include humans "over-the-loop", giving the user an option to intervene when they deem it necessary while not having user feedback be an execution requirement. With this strategy, we hope to increase the accuracy of solutions with minimal losses in execution time. This paper describes our vision of this strategy and associated problems.