A Survey of Algorithm Debt in Machine and Deep Learning Systems: Definition, Smells, and Future Work

2026-04-07

AI Summary

This study addresses the performance degradation and limited scalability in machine learning and deep learning systems caused by algorithm debt. It presents the first systematic conceptualization of algorithm debt, drawing on a comprehensive literature review of 42 core studies. Through qualitative analysis and categorical synthesis, the work rigorously defines algorithm debt, delineates its key characteristics, and identifies representative “bad smells.” By bridging a critical gap in technical debt research at the algorithmic level, this paper establishes a theoretical foundation for detecting, identifying, and mitigating algorithm debt, while also outlining a roadmap for future research in this emerging area.

Technology Category

Machine Learning: Deep Learning AlgorithmsData Mining & Knowledge Management: Scalability, Parallel & Distributed SystemsCognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

Responsible Web: Human-perceived consequences of algorithmic deployment on the webGraph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphsWeb Mining and Content Analysis: Web data quality in the era of algorithmically-generated content

Abstract

The adoption of Machine and Deep Learning (ML/DL) technologies introduces maintenance challenges, leading to Technical Debt (TD). Algorithm Debt (AD) is a TD type that impacts the performance and scalability of ML/DL systems. A review of 42 primary studies expanded AD's definition, uncovered its implicit presence, identified its smells, and highlighted future directions. These findings will guide an AD-focused study, enhancing the reliability of ML/DL systems.

Problem

• Algorithm Debt
• Technical Debt
• Machine Learning
• Deep Learning
• Software Maintenance

Innovation

• Algorithm Debt
• Technical Debt
• Machine Learning Maintenance
• Deep Learning Systems
• Code Smells